research paper stock market

• Browse All Articles
• Newsletter Sign-Up

FinancialMarkets →

No results found in working knowledge.

• Were any results found in one of the other content buckets on the left?
• Try removing some search filters.
• Use different search filters.

SYSTEMATIC REVIEW article

Covid and world stock markets: a comprehensive discussion.

• 1 Department of Management Sciences, Lahore College for Women University, Lahore, Pakistan
• 2 Department of Management Sciences, National University of Modern Languages, Islamabad, Pakistan

The COVID-19 outbreak has disturbed the victims' economic conditions and posed a significant threat to economies worldwide and their respective financial markets. The majority of the world stock markets have suffered losses in the trillions of dollars, and international financial institutions were forced to reduce their forecasted growth for 2020 and the years to come. The current research deals with the impact of the COVID-19 pandemic on the global stock markets. It has focused on the contingent effects of previous and current pandemics on the financial markets. It has also elaborated on the pandemic impact on diverse pillars of the economy. Irrespective of all these destructive effects of the pandemic, still hopes are there for a sharp rise and speedy improvement in global stock markets' performance.

Introduction

The world is experiencing the worst health and economic disaster in the shape of COVID-19 pandemic. Dealing with this pandemic is the most challenging task being faced by human beings since the Second World War ( Maqsood et al., 2021 ). Coronavirus has pushed the markets toward the danger zone. The market panic has been started. This disease is contagious even before it shows obvious symptoms. It is quite difficult to hold people in quarantine in this outbreak. That's the narrative, and we haven't gotten very far into it yet. So, the potential for market disruption because of a scary narrative is quite high.

—Robert James Shriller, Nobel Memorial Prize Winner in Economic Sciences, 2013.

The epidemiological perspectives are not required to be understood here. Currently, well-informed individuals ought to have some know-how about the basics of contagious diseases. Times of fear are also times of rumor and misinformation; knowledge is the antidote ( Baldwin et al., 2020 ). The COVID-19 outbreak was officially reported in the Wuhan City of China in December 2019 and covered all continents of this globe other than Antarctica ( Hui et al., 2020 ). COVID-19 is a distinctive black swan event, and we are unaware of its existence, expansion, breadth, depth, magnitude, and even its disappearance ( He P. et al., 2020 ; He Q. et al., 2020 ). World Health Organization (WHO) officially declared COVID-19 a pandemic on 11th March 2020 ( Cucinotta and Vanelli, 2020 ). The pandemic has severely hit global economies ( Shafi et al., 2020 ). It has disrupted the life and lifestyle of almost everyone ( Aqeel et al., 2021 ). Almost no one has been left untouched. Another pandemic of information and misinformation is keeping pace with it during this pandemic, spreading fear and anxiety ( Koley and Dhole, 2020 ). The outbreak has changed the outlook of this globe within no time at all. Human beings are struggling with the long-lasting effects of this disease and the unforgettable reality of their existence which has never happened before. The pandemic affected more than 107 million people, with around 2.3 million causalities, and the numbers of cases are escalated day by day. The alarming point is the growth factor of this disease, where 100 contaminated cases create another 10,000 within a very limited time ( Bagchi et al., 2020 ).

The people from this generation have seen wars. They have seen the collapse of the Soviet Union. They have seen extremely dangerous terrorist attacks. They have seen the burst of financial bubbles, and they have seen the effects of climate change. However, they had not seen anything like the coronavirus before. A similar case has not existed for more than one hundred years. They were not ready for it, and they did not know how to respond to it. Since it was something that no one had any prior experience with, the pandemic has also led to reconsidering some things which were always previously thought either right or wrong ( Sharma, 2021 ).

The COVID-19 pandemic has spread globally, has made millions of people sick, and triggered an international response spearheaded by the World Health Organization to stop its spread. From Wuhan, China, it spread like wildfire. The virus has now visited almost every nation in the world, bringing helplessness and death with it. None are spared, and in some way or another, almost everyone has become a victim. In a recent message, the WHO warned that the worst is yet to come. The coronavirus has not only triggered disease and death, and it has affected almost every aspect human life. There is a long list of disruptions to daily life in the cities and states with lockdown, global sporting events, weddings, social events, post-poned ceremonies; all this has elicited the global crisis. Moreover, industries worldwide have been affected; stock markets have been reported in record downfall; airlines, travel, tourism, and hospitality sectors are the major victims of this pandemic. A significant disaster is job loss in various sectors ( Koley and Dhole, 2020 ).

Crucial and groundbreaking strategies are required to protect not only human lives but also to safeguard economies and uplift economic growth and financial health. Nations are exposed to a global health crisis, the like of which has not occurred for a century. This crisis is killing human beings, enhancing human distress, and upsetting the lives of individuals. This can be considered a sort of human, social, and economic crisis ( Mishra, 2020 ). The best efforts by governments from every country have failed to halt its spread: cities were put under lockdown; people were advised to stay at home; international borders were closed; travel bans at local, national and international level were imposed; markets, schools, universities and shopping complexes were closed. Quarantine and self-isolation have been advised to stop the spread of COVID-19. The virus has triggered an unprecedented global crisis which led the WHO to provide technical guidance for government authorities, healthcare workers, and other key stakeholders to respond to community spread ( Koley and Dhole, 2020 ).

In the intermingled economies, the Covid-19 pandemic came as a global distress that affects both the demand and supply side concurrently. Rapidly growing infectivity limits labor supply and badly affects productivity, whereas supply disruptions are also caused by social distancing, lockdowns and industry closures. On the other hand, disruption on the demand side is caused by reduced consumption, unemployment, and income loss and these economic prospects result in reduced company investment. The unpredictability about the path, instance, enormity and impact of Covid-19 could create a vicious cycle of redundancy, less consumption, and business closures, leading to financial distress. To identify and determine this extraordinary shock is the key challenge for the experiential analysis of this pandemic. The unprecedented nature of COVID-19 makes it difficult to recognize its non-linear effects, cross-country spillovers, and quantify unobserved factors to compose forecasts ( Chudik et al., 2020 ).

International institutions including the FAO, ILO, IFAD, and WHO jointly declared this pandemic a global challenge to food systems, public health, trade, and industry. Overwhelming social and economic disruptions put tens of millions of individuals at risk of falling below the poverty line. According to another approximation, by the end of this year, the number of undernourished people could increase by up to 32 million, which are ~690 million at present. It also poses an existential threat to a considerable number of business ventures. The world has a 3.3 billion workforce and ~50% of which are near to being unemployed. Significant individuals are informal workers with limited access to productive assets, quality health, and the majority lack social protection. Due to lockdowns, they lost their means to earn money and became incapable of feeding themselves and their families because for most their daily food depends upon their daily wages earned. Such a devastating effect on the entire food chain has exposed the vulnerability of this pandemic. Farmers have no access to markets, nor can buy inputs or sell their output and result in a reduced harvest. In addition to market shutdown, trade limitations, border closures, and detention measures dislocate food supply chains nationally and internationally, which badly influenced a healthy diet. Small scale farmers are the soft target of COVID-19 and placed nutrition and food security of the most marginalized population under threat, as income producers fall ill, die, or otherwise lose their work ( Chriscaden, 2020 ).

It is still difficult to understand the recovery due to the development of vaccines. To understand corona's economic impact, the following charts and maps exhibit real statistics so far.

Impact on Jobs

A report published by the OECD (2020) shows the impact of COVID-19 and containment measures on OECD economies where people were prohibited from going to work, resulting in a significant drop in business activity and extraordinary job losses. In some countries, millions have been moved to reduced hours and most people worked up to ten times fewer hours. Moreover, the rate of entire job loss is also very high. Some people are more exposed to this pandemic than others. As young people and women workers are at greater risk due to less secured and unskilled jobs. They are also associated with the industries most affected by this unprecedented shock, including restaurants, cafés and tourism.

Causing Recession

Worldwide economic downturn caused by the COVID-19 pandemic forced the Organization for Economic Cooperation and Development (OECD), International Monetary Fund (IMF), and World Bank (WB) to revise their forecasts and reported a significant decline in the projected rate of growth in late 2019 and mid-2020. Such deterioration can be seen in the IMF figures in which global economic growth forecasts declined from +3.4% to −4.4% during October 2019 and October 2020. In the same way, OECD also revised its forecast and lowered the growth rate from positive 2.9% in November to −4.5% in September 2020. In June 2020, OECD anticipated the blow of another wave of infections.

Impact on Travel

The travel industry is one of those acutely damaged industries due to lockdowns, border closures, and abandoned flight operations. Airlines are not only canceling flights, but customers also restrict themselves from holidays and business trips. A recently discovered subsequent wave of COVID-19 has forced national and international airlines to promulgate new travel restrictions and tighten their policies. While providing data of 2020, Flight tracking service Flight Radar 24 reveals a huge hit in number of flights worldwide and requires a long way to recover ( Jones et al., 2021 ).

Impact on Tourism

Tourism is another badly affected industry due to this unprecedented pandemic. The World Tourism Organization, also known as UNWTO (2020) marked this pandemic as a serious threat to the travel and tourism sector. Many jurisdictions put restrictions on international travel to restrict the spread of the virus; some fully closed their borders, resulting in a massive decline in demand. In 2020, tourism reported a loss of ~1 billion tourists, equivalent to US$ 1.1 trillion in international tourism receipts. This decline in international tourism could cause an ~$2 trillion loss in global GDP, over 2% of the global GDP in 2019. While predicting a rebound in the global tourism industry, UNWTO presented an extended scenario for the year 2021 to the year 2024. According to them, global tourism will start recovery from the second half of the year 2021 but it will take 2.5 to 4 years to return to 2019.

Impact on Stock Markets

The capital markets are at the front line of any country's economy, and the stock markets are considered the indicator of any economy ( He P. et al., 2020 ; He Q. et al., 2020 ). The COVID-19 outbreak has disturbed the victims' economic conditions and posed a significant threat to the worldwide economies and their respective financial markets ( Barro et al., 2020 ; Ramelli and Wagner, 2020 ). The majority of the world stock markets have suffered in terms of trillion-dollar losses ( Lyócsa et al., 2020 ) and international financial institutions were forced to reduce their forecasted growth for 2020 and the years to come ( Boone et al., 2020 ). The root cause of this severe decline is the exposure of stock markets to several risks, for instance, the global financial crisis of 2008, which had pushed these markets in a melting position ( Dang and Nguyen, 2020 ). The current pandemic has affected the global stock markets significantly compared to the SARS virus, which was spread in 2003 as China has got tremendous development in comparison to the last 17 years and recognized as a leading economy of the world and also a global production hub, manufacturing the highly demanded technology products ( Alameer et al., 2019 ).

Effects of Previous Pandemics on Stock Markets

Scholars have argued that previous pandemics triggered fragile stock markets ( Chen et al., 2018 ) and impeded stock market participants' decision-making capacity by reducing their active involvement in stock market trading ( Dong and Heo, 2014 ). The literature has provided empirical evidence of the stock market reactions to significant systematic events. The research has shown the cyclical nature of the stock market reactions and the factors that affected the stock markets ( Keating, 2001 ). The historical performance of stock markets has been documented in the previous literature regarding influenza and other major epidemics. Similarly, the scholars have examined the influences of significant events on the stock markets, i.e., Severe Acute Respiratory Syndrome (SARS), ( Chen et al., 2018 ), natural disasters ( Caporale et al., 2019 ), corporate events ( Ranju and Mallikarjunappa, 2019 ), public news, and political events ( Bash and Alsaifi, 2019 ). Some other studies have also demonstrated that SARS in 2003 weakened the Taiwanese economy ( Chen et al., 2007 ) and regional stock markets ( Chen et al., 2018 ).

The previous studies have comprehensively examined the association between outbreaks and stock market performance. Kalra et al. (1993) investigated the disaster of the Soviet Chernobyl nuclear power plant. Delisle (2003) recognized that effects were of greater intensity after SARS (2003) than the Asian financial crisis. Nippani and Washer (2004) investigated the effects of SARS on global financial markets and found that it influenced the markets of China and Vietnam. Lee and and McKibbin (2004) reported the strong effect of SARS on human beings and financial integration. Loh (2006) explained a robust linkage between SARS and airline stocks performance in Canada, China, Hong Kong, Singapore, and Thailand and illustrated that the stocks of the aviation sector are more sensitive than non-aviation stocks. MckKibbin and Sidorenko (2006) investigated the influenza epidemic's impact on the global economy's growth by considering its diverse magnitudes like slight, moderate and intense. Moreover, Chen et al. (2007) noticed the negative effects of SARS on the hotel industry's stock prices in Taiwan. They also investigated the significant influence of SARS on the four major stock markets of Asia and China. Nikkinen et al. (2008) discovered the impact of the 9/11 incident on the global stock prices; however, the markets recovered rapidly. Al Rjoub (2009) also studied the influence of financial crisis on stock market.

Besides, Kaplanski and Levy (2010) studied the effect of aviation accidents on stock returns and established that price fluctuations are sensitive to such incidents. Al Rjoub (2011) and Al Rjoub and Azzam (2012) investigated the impact of the Mexican tequila crisis (1994), Asian-Russian financial crisis (1997–98), 9/11 incident, Iraq war (2004), financial crisis (2005), and global financial crisis (2008–09) on the stock compensation behavior in Jordan's Stock Exchange. Righi and Ceretta (2011) established the positive effect of the European debt crisis (2010) on European markets' risk aptitude, especially the German, French, and British markets. Schwert (2011) explored the variabilities in the prices of US stocks during the financial crisis. Mctier et al. (2011) found the negative impact of Flu on the intensity of trading activities and stock returns in the USA. Besides, Rengasamy (2012) examined the effect of Eurozone sovereign debt-related policy announcements, development rewards, and stock market volatility on Brazil, Russia, India, China, and South Africa. Karlsson and Nilsson (2014) found the negative impact of the 1918 Spanish flu epidemic on capital returns. Lanfear et al. (2018) conducted a study to explore the effect of cyclones on stock returns, and they observed the effect of emergencies on stock returns. Chen et al. (2018) examined the influence of SARS on Asian financial markets.

Brief Overview of Literature

Studies have elaborated on the performance of global stock markets affecting the COVID-19 outbreak ( Ahmar and del Val, 2020 ; Al-Awadhi et al., 2020 ; Liu et al., 2020 ; Zhang et al., 2020 ). The pandemic has decreased investors' confidence level in the stock market as the market uncertainty was very high ( Liu et al., 2020 ). Iyke (2020) explained that COVID-19 has robust and continual negative effects on the global economy. Ahmar and del Val (2020) used ARIMA and SutteARIMA and forecasted the short-term impact of COVID-19 on Spain's IBEX index. They further explained that SutteARIMA is the better statistical measure in forecasting such impact.

Moreover, Alam et al. (2020) explained that pandemic has greatly hit Australia's capital market right from the start of 2020. The stock market has shown a bearish trend, though some sectors were at high risk and others have performed well. The researchers have focused on initial volatility and sectoral returns in eight different sectors. They have analyzed the data using the event study method and 10-days window for the official announcements of COVID-19 events in Australia. The findings revealed that some sectors performed well on the day of the announcement. Simultaneously, some others also showed good performance after the announcement except the transportation sector, which performed poorly.

The pandemic has posed severe challenges to the global economies ( Wang et al., 2021 ) and it has also created mental health issues ( Abbas et al., 2021 ). Chowdhury et al. (2020) examined the impact of COVID-19 on economic activities and stock markets worldwide. The study has targeted 12 countries from four continents from January-April 2020 by using the panel data. The stock market impact was measured using the event study method, and economic impact was measured using the panel vector autoregressive model. The results showed the extremely negative effects of pandemic variables on stock returns. Singh et al. (2020) investigated the influence of COVID-19 on the stock markets of G-20 states. The study used an event study for measuring abnormal returns and panel data to describe the causes of abnormal returns. The data consisted of 58 days of post-COVID period news provided by international media and 120 days before the event. The findings exhibited the significant negative abnormal returns during the event days. Liu et al. (2020) also examined the pandemic's effect on the most affected countries' stock markets by using the event study. The researchers revealed the negative effects of COVID-19 on the stock markets' performance.

He P. et al. (2020) and He Q. et al. (2020) also used the event study method to explore the impact of COVID-19 on Chinese industries and stock market performance. It has been observed that some industries were severely affected by the pandemic (mining, environment etc.). However, some other industries have faced limited effects of an outbreak (manufacturing, education etc.). Machmuddah et al. (2020) used the event study method to observe consumer goods' share prices before and after COVID-19. The data about daily stock prices and stock trade volume has been collected before and after the pandemic. Significant differences have been observed between daily closing prices and stock trade volume before and after the pandemic. Liu et al. (2020) used an event study method to study the short term impact of the outbreak on the stock market indices of 21 countries strongly affected by pandemic (Italy, UK, Germany etc.). Asian countries have taken the severe negative effect of the pandemic as compared to other states. Khatatbeh et al. (2020) also applied the event study method to discover the impact of COVID-19 on some targeted countries' stock indices by employing the daily stock prices and found a significant negative impact on returns.

Al-Awadhi et al. (2020) investigated the association of pandemic and stock market outcomes in the Chinese stock market. The findings showed the effect of pandemic cases and deaths on the stock returns of different organizations. Baker et al. (2020) claimed that COVID-19 strongly affects the US stock market compared to previous epidemics, including the Spanish Flu. Eighteen market jumps were observed from February-March 2020. The market jumps were considered to be the largest ones since 1990. The causes behind such jumps were the lockdowns and production cut. Ozili and Arun (2020) described that COVID-19 uncertainty and the fear of losing profit have resulted in 6 trillion USD in the global stock market on 24th February 2020. Similarly, the S&P 500 index has faced a loss of 5 trillion US dollar. The research also demonstrated the significant influence the pandemic on the opening, highest, and lowest stock indices in the US. Ngwakwe (2020) illustrated the influence of COVID-19 outbreak on some targeted stock indices (SSE, Euronext, and DJIA) by collecting the data for 50 days before and 50 days within the pandemic. The differential effects of the pandemic were observed in different stock markets. DJIA stock returns were decreased, SSE increased, however, S&P 500 index and Euronext 100 revealed insignificant effects.

He P. et al. (2020) and He Q. et al. (2020) examined the direct effects of COVID-19 spillovers on the stock market. The daily return data has collected from China, Italy, South Korea, France, Spain, Germany, Japan and the USA. The findings showed the negative short term effects of COVID-19 on the stock indices. Zhang et al. (2020) elaborated the impact of pandemic fear on the pattern of systematic risk and country-specific risk in the global financial markets. They explained the volatile nature of financial markets and the huge impact of uncertain market conditions on financial market risk. Sobieralski (2020) evaluated the effect of COVID-19 on employment and the aviation industry. The stock returns of China and US stocks have declined at a record level. Qin et al. (2020) investigated the influence of outbreak on oil markets.

Sansa (2020) explained the association between COVID-19 recorded cases and financial markets systems of SSE and DJIA during March 2020. Aslam et al. (2020) studied the impact of COVID-19 on 56 global stock market (developed, developing, emerging, and frontier) indices by using the network method. Topcu and Gulal (2020) have discovered a huge impact on Asian markets as compared to European markets. Ashraf (2020) explained that confirmed cases more strongly affect the stock market than deaths. Czech et al. (2020) used the TGARCH model and found the negative impact of COVID-19 on Visegrad stock market indices. They discovered that stock markets were seriously affected when the disease's nature was changed from epidemic to pandemic.

Zhang et al. (2020) also investigated the influence of COVID-19 on the stock markets of 10 countries. It was concluded that European stock markets showed connectivity during the outbreak; however, US markets could not show a leading role before and during the pandemic. Okorie and Lin (2021) discovered the occurrence of financial contagion during the pandemic. Corbet et al. (2020) presented some interesting insights. They illustrated that pandemic greatly affected the companies having names related to the virus, although these companies were not related to the virus.

The current research work basically pertains to the comprehensive discussion about the past present and future of world stock markets. For the sake of achieving the research aims, it has also presented a somehow brief yet inclusive debate about the happenings in the renowned stock markets. It has focused on the major market indices belong to different regions and also it has attempted to explain the actual position of some famous indices with the help of underlying real time data based graphs. Its major contribution is presenting the diverse opinions of traditional and behavioral finance regarding the behavior of stock market participants.

A General Debate About Stock Markets Performance

The global stock markets have been reported for their record decline. On 23 March 2020 the S&P 500 Index witnessed an usual drop of 35% compared to the record high on 18 February 2020. In no time at all, the intensity of this record fall became comparable with the financial crisis of 2008, black Monday of 1987, and the great depression of October-November, 1929 ( Helppie McFall, 2011 ). Fernandes (2020) also explained that the US S&P 500 index went down to 30% during March 2020. He further described that the UK and Germany's stock markets were noticed for their worst performance than the US market. The returns of these two markets were fallen by 37 and 33%, respectively. However, the worst performers in the global stock markets were Brazil (−48%) and Columbia (−47%).

Japan's market index dropped more than 20% compared to the record high values of December 2019. S&P 500 Index and Dow Jones share points were declined by 20% in March 2020. The Nikkei Index also reported the same downfall. The Colombo Stock Exchange witnessed a 9% drop in share value and experienced three market halts during mid-March 2020. The Indonesian stock market followed a similar decline. In April 2020, the index was opened with a 64.06 points decline. The UK-FTSE index plunged by 29.72%. The DAX (Germany) index was dropped by 33.37%, CAC (France) by 33.63%, NIKKEI (Japan) by 26.85%, and SUNSEX (India) want down by 17.74% ( Machmuddah et al., 2020 ). Shanghai Composite went down to 2,660.17 points on 23rd March 2020, showing a decline of 12.49% compared to December 2019. KOSPI touched the peak level of 2,204.21 points on 27th December 2019 and dropped to the lowest point of 1,457.64 on 19th March 2020, showing a drop of 33.87%. The BSE SENSEX reported the highest points of 41,681.54 on 20th December 2019. BSE SENSEX plunged to 25,981 points on 23rd March 2020 due to the COVID-19 outbreak, demonstrating a decline of 37.66%. FTSE 100 showed an upward trend on 27th December 2019 with a record index of 7,644.90 points, but it reflected the downward trend followed by a pandemic with an index value of 4,993.89 34.67% decline. The NASDAQ 100 Index reached 8,778.31 points on 26th December 2019 and observed the negative effects of the COVID outbreak by touching 7006.92 points with a declining trend of 20.17%. Moreover, MOEX revealed a bullish trend on 27th December 2019 with an index value of 3,050.47 points and reflected the effects of COVID-19 by reaching 2,112.64 points with the corresponding decline of 30.74%.

Besides, FTSE MIB reached the record level of 24,003.64 points on 20th December 2019 and then touched 14,894.44 points due to pandemic on 12th March 2020 with a declining rate of 37.94%. Nikkei 225 demonstrated an upward trend with the peak value of 24,066.12 on 17th December 2019 and represented the lowest range of 16,552.83 points following the pandemic on 19th March 2020 with the corresponding decline of 31.21%. CAC 40 represented 6,037.39 points on 27th December 2019, consequently faced the sharp jerk of 37.80% on 18th March 2020. DAX exhibited an ascending trend on 16th December 2020 with a peak value of 13,407.66, with the corresponding decline of 8,441.71 on 18th March 2020, signifying an increase of 37.04%. Moving forward, S&P/TSX jumped to 17,180.15 on 24th December 2019 and showed the devastating effects of COVID-19 with the sharp decline of 34.64% on 23rd March 2020. Besides, FTS/JSE reflected 3,513.21 points on 20th November 2020 and affected by the outbreak with a decline of 36.37% on 23rd March 2020 (Investopedia).

However, the global stock markets regained and demonstrated a bullish trend during the days of April 2020. The S&P 500 index increased by 29% and regained the strong position it had held in August 2019 ( Cox et al., 2020 ). Shanghai Composite index further increased by 8.22% in May 2020. KOSPI index showed a bullish trend and the index increased by 27.05%. Similarly, BSE SENSEX recaptured its position and touched 33,717.62 points on 30th April 2020, representing the rise of 22.94%. FTSE 100 secured an 18.33% increase, and the index targeted 6,115.25 points on 29th April 2020. NASDAQ 100 touched 9,485.02 points on 20th May 2020 with the respective rise of 26.12%. MOEX showed an upward trend with a 74.64% increase on 13th April 2020. Also, BOVESPA regained by 23.56% on 29th April 2020 and touched 83.170.80 points. FTSE MIB upbeat and reached 18,067.29 on 29th April 2020. On 20th May 2020, NIKKEI Index climbed at 20,595.15 points, reflecting an increase of 19.62%. Moreover, CAC 40 revived by 19.61% on 29th April 2020. DAX invigorated with the 24.79% increase on 20th May 2020. S&P/TSX touched 15,228 points on 29th April 2020, FTSE/JSE recovered by 27.09% on 20th May 2020, beating the outbreak's negative effect (Investopedia).

The stock market indices worldwide have been categorized in terms of Major Stock Indices, Global Stock Indices, and World Stock Indices etc. The Major World Stock market indices as well as their respective countries have been presented in the Table 1 .

Table 1 . Major world market indices.

Graphical Representation of Some Leading Indices

Source of all figures: tradingeconomics.com .

Figure 1 represents the stock market performance of the S&P ASX 50 index of Australia. It can be seen that the index was performing well-during January 2020, when COVID-19 was at its initial phase. However, March seemed to be a nightmare, when the index plunged and reached the lowest level as COVID-19 spread rapidly and hit a majority of the nations. But the index revived during April 2020, and a gradually limited bullish trend was observed. In-spite of such revival, the index could not reach its peak as the world is still facing the 3rd wave of the pandemic.

Figure 1 . S&P ASX 50 (Australia). Source: tradingeconomics.com . Reproduced with permission.

Figure 2 exhibits the stock market conditions of DAX Germany. The stock market did perform well-until February 2020, it showed a bearish trend in March 2020, followed by a gradual increase, and finally, it realized the position as it was before the pandemic.

Figure 2 . DAX (Germany). Source: tradingeconomics.com . Reproduced with permission.

Figure 3 demonstrates the stock market situation of Dow Jones Industrial Averages, one of the USA's leading indices. The same situation was observed just like previous indices. The bullish trend was observed before March 2020, followed by the bearish trend during March-April, 2020. Index regained slowly, and revival leads to the extreme upward movements.

Figure 3 . Dow Jones industrial averages (USA). Source: tradingeconomics.com . Reproduced with permission.

Figure 4 illustrates the stock market trend of CAC 40, the index of France. The index was at its peak during February 2020. However, a sudden jerk was observed during March 2020, and the index touched the lowest points. The index recovered quite slowly, and to date, it could not recover its previous position. The fluctuations in the index can still be noticed.

Figure 4 . CAC40 (France). Source: tradingeconomics.com . Reproduced with permission.

The variations in the FTSE 100 index of Europe's market conditions can be seen in Figure 5 . The bullish trend can be observed before March 2020, followed by the extreme bearish trend. The index went to the historical lowest points during March 2020. The upward movements were started during April 2020; however, slow movements were there, and the index is still in a slow recovery phase.

Figure 5 . FTSE-100 (Europe). Source: tradingeconomics.com . Reproduced with permission.

SENSEX index is a famous stock market index of India. Figure 6 is representing its performance. In January 2020, though the index was not performing very well, it faced the effects of COVID-19 during March 2020. The extreme slow revival was observed after March, and the index remains at the same pace. However, the gradual upward trends lead the index to its highest peak in 2021, as shown in the figure.

Figure 6 . SENSEX (India). Source: tradingeconomics.com . Reproduced with permission.

Figure 7 depicts Japan's famous index, i.e., the Nikkei 225. The index was in the recovery phase during January 2020; however, a bearish trend was observed from the outbreak. After March 2020, the recovery phase was there, but static movements were observed. Nevertheless, these slow recoveries finally touched the highest peak in 2021.

Figure 7 . Nikkei-225 (Japan). Source: tradingeconomics.com . Reproduced with permission.

Figure 8 deals with the NASDAQ stock market performance, one of the USA's leading indices. During the start of 2020, its performance was below average, ultimately reaching the lowest points in March 2020 as per the COVID-19 effects. The index escalates gradually, and to date, it jumped and touched the peak level.

Figure 8 . NASDAQ (USA). Source: tradingeconomics.com . Reproduced with permission.

Referring to Figure 9 , the PSX-100 of Pakistan was performing well-before the sharp rise of COVID-19 in Pakistan. However, the month of March 2020 proved to be a terrible one; the index plunged and touched the lowest level. A slow revival was observed, which ultimately hit the highest points during 2021, as showing in the Figure 9 .

Figure 9 . PSX-100 (Pakistan). Source: tradingeconomics.com . Reproduced with permission.

The performance of the S&P 500 index, a prominent index of the USA, seems to be similar to the NASDAQ stock market index. However, before COVID and the recovery phase after March 2020 is better than the NASDAQ index. Currently, the index has reached the highest level as shown in Figure 10 .

Figure 10 . S&P 500 (USA). Source: tradingeconomics.com . Reproduced with permission.

Figure 11 exhibits the Shanghai Stock Exchange Composite index of China, the origin of the COVID-19 pandemic. The SSE index is the outperformer index of China; however, it was severely affected by the pandemic. The index plunged from the start of the outbreak up to June 2020, followed by a sharp rise and now, with the gradual increase, the index has reached the maximum points.

Figure 11 . SSE composite (China). Source: tradingeconomics.com . Reproduced with permission.

Implications for Stock Market Participants

The current study has some implications for market participants and policymakers, i.e., investors, managers, corporations, and governments. The investors must have market know how to invest their resources in favorable avenues even during the contingent market conditions, which have just happened during COVID-19. The investors can take guidance from the mangers and policymakers in this regard. In this way, investors can take rational decision making for their investments. Moreover, investors can generate their portfolios and risk management strategies. Besides, investors must focus on diversification to avoid losses during the pandemic situation.

The managers are the key stakeholders of financial markets; they have experience with stocks' risky nature during the pandemic and can take preventive measures accordingly. The mangers can increase the confidence level of investors, which lead them to make long term investments.

Governments can also play a vital role in assisting with the outbreak in tax rebates and interest-free loans. Governments can even facilitate the national markets by relaxing the lending policies and providing short-term loans on relaxing terms. Governments can conduct surveys and can assist investors in reducing their uncertainty.

Policymakers can develop successful methodologies for balancing financial investments during the outbreak. For this, they can focus on understanding the dynamics of stock markets in devising effective strategies. Besides, policymakers can integrate policies to cope-up with the financial and economic impacts of the COVID-19 outbreak. The emphasis must be on the improvement of stock market stability.

Unlocking the Future in Post-COVID-19 World

The aim of realizing sustainable growth, a big challenge for all economies, has been underscored by the novel virus. The pandemic has directed that economies are not goal-oriented in terms of their aspiration; they are required to achieve the milestones of a robust global economy. The greatest accomplishments are always achieved through the heights of determination. History is always there to provide lessons for the future. Nearly 75 years ago, amid World War II and one of Britain's most difficult hours, Winston Churchill inspired the whole nation not with the slogans to “reconsider what is achievable” but with a firm determination to “never surrender.” Now at the most difficult time of the century, we are required to continue our fight for what the world needs instead of reconsidering the sustainable development goals. This crisis requires a determined global effort to “build back better” by making a big reset to reach where we were before ( Kharas and and McArthur, 2020 ). Moreover, the leaders of the world must draw a new course of action for improving the functioning of international financial and monetary system to make it strong enough to cope with any such crisis in future ( Coulibaly and Prasad, 2020 ). The stock markets had to face the worst situation in the last 30 years, business operations have abruptly failed, and various economic sectors have been critically affected. However, the best point that came out of the COVID-19 pandemic is the businesses' pressure to be innovative and redefine their operations. One example is the tech community, which has been progressing to facilitate the community in adopting the technology to deal with the pandemic's challenges. Such technological innovations assist specific divisions of organizations or even the whole organizations to carry on their operations irrespective of the current contingent situation. Certainly, the world and the multinational business models will face diverse post-virus issues. Following the COVID-19 pandemic, the nations will observe new policies relating to restructuring and operational strategies, i.e., strategic workforce planning including remote staff planning, flexible conventions, workers proficiency, best practices and HR strategies; Crisis response and business continuity planning, risk control strategies and measures; Financial resources to weather future unforeseen events; Cloud-enabled IT infrastructure (and the attendant improved cybersecurity procedures); and the Redundant sourcing of necessities (inventory, materials and individuals), ( David, 2020 ).

The prospects of the stock markets and the economies are based on the availability and accessibility of the vaccine. The optimism about the vaccine has revitalized the investors' appetite regarding hotels, energy firms, and airlines. However, some others have been brutally affected by the pandemic and are forced to sell their respective market shares. Stock markets are largely dealing with the sentiment that tomorrow may be better than today, leading to a fundamental and perhaps enduring sea change. The development of more vaccines would pave the way for more optimism. What this result demonstrates is that while the virus is not yet beaten, it is beatable. That ray of light has lit up stock markets around the world. As usual, some stock market participants are there to look for something else to worry about ( Jack, 2020 ).

The current study deals with the impact of the COVID-19 pandemic on the global stock markets. It has focused on the contingent effects of previous and current pandemics on the financial markets. It has also elaborated on the impact of the pandemic on diverse pillars of the economy. The pandemic has severely hit the worldwide markets and posited challenges for economists, policymakers, head of states, international financial institutions, regulatory authorities, and health institutions to deal with the long-lasting effects of the outbreak. It has opened our eyes to concentrate our efforts to protect the future health of citizens and the also financial issues. In the current pandemic situation, the stock markets faced the effects of Covid-19, and this back-and-forth is ongoing. The majority of the world stock markets have suffered trillion-dollar losses ( Lyócsa et al., 2020 ). International financial institutions like IMP and World Bank have been forced to reduce their forecasted growth for 2020 and the years to come ( Boone et al., 2020 ). The global stock markets have been reported for their record decline. The month of March 2020 saw an unusual drop in most worldwide indices like the S&P 500 Index, NASDAQ, NIKKEI, SSE composite, CAC-40; DAX etc. However, the global stock markets regained and demonstrated the bullish trend during the days of April 2020. Irrespective of all these cyclical effects of the pandemic, still hopes are there for the sharp rise and speedy improvements in global stock markets' performance. Moreover, these past events have become a key for mankind to get insights for better future planning ( Su et al., 2021 ).

Behavioral vs. Conventional Finance

The two polar aspects of finance i.e., traditional finance and behavioral finance have also shed light on the psychology of investors during the COVID-19 pandemic. As far as traditional finance is concerned, investors behave rationally. The rational attitude of investors restricts them from imitating the decisions of others. Investors get the basic facts and figures about the stock markets through their own efforts, resultantly the fear of avoiding future losses compel the investors to sell their stocks and the market shows bearish trend ( Jabeen and Rizavi, 2021 ). The same has happened in the world's stock markets during the peak of pandemic. There have been sudden jerks observed around the global stock markets ( Jabeen and Farhan, 2020 ).

On the other hand, behavioral finance is supposed to consist of a set of theories which focus on the irrationality of investors. The viewpoint of irrationality of investors is the foundation of behavioral finance. The irrational aspect of investors forces them to follow the decision of other investors by setting aside their own information. In such context, investors have confidence on the decision of other investors as they feel that others may possess better information skills ( Jabeen and Rizavi, 2021 ). As a result the panic market conditions lead investors to blindly follow the others to protect their investment and market also depicts the bearish trend, the one which has been seen during the COVID-19 outbreak.

This debate has proven that both the traditional finance and behavioral finance have provided the same mechanism during the COVID-19 pandemic, irrespective of the fact that these pillars of finance deal with the opposing behaviors of investors i.e., rational and irrational. In both of the scenarios, the investors have sale their shares and resultantly the bearish trend has been observed.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Author Contributions

All authors listed have made a substantial, direct and intellectual contribution to the work, and approved it for publication.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Abbas, J., Wang, D., Su, Z., and Ziapour, A. (2021). The role of social media in the advent of COVID-19 pandemic: crisis management, mental health challenges and implications. Risk Manag. Healthc. Policy 14, 1917–1932. doi: 10.2147/RMHP.S284313

PubMed Abstract | CrossRef Full Text | Google Scholar

Ahmar, A. S., and del Val, E. B. (2020). SutteARIMA: short-term forecasting method, a case: Covid-19 and stock market in Spain. Sci. Total Environ. 729:138883. doi: 10.1016/j.scitotenv.2020.138883

Al Rjoub, S. A. (2011). Business cycles, financial crises, and stock volatility in Jordan stock exchange. Int. J. Econ. Perspect. 5:21. doi: 10.2139/ssrn.1461819

CrossRef Full Text | Google Scholar

Al Rjoub, S. A., and Azzam, H. (2012). Financial crises, stock returns and volatility in an emerging stock market: the case of Jordan. Journal of Economic Studies 39, 178–211. doi: 10.1108/01443581211222653

Al Rjoub, S. A. M. (2009). Business cycles, financial crises, and stock volatility in Jordan stock exchange. Soc. Sci. Electron. Publish. 31, 127–132.

Google Scholar

Alam, M. M., Wei, H., and Wahid, A. N. (2020). COVID-19 outbreak and sectoral performance of the Australian stock market: an event study analysis. Curr. Protocols 60, 482-495. doi: 10.1111/1467-8454.12215

Alameer, Z., Elaziz, M. A., Ewees, A. A., Ye, H., and Jianhua, Z. (2019). Forecasting copper prices using hybrid adaptive neuro-fuzzy inference system and genetic algorithms. Nat. Resour. Res. 28, 1385–1401. doi: 10.1007/s11053-019-09473-w

Al-Awadhi, A. M., Alsaifi, K., Al-Awadhi, A., and Alhammadi, S. (2020). Death and contagious infectious diseases: impact of the COVID-19 virus on stock market returns. J. Behav. Exp. Finance 27:100326. doi: 10.1016/j.jbef.2020.100326

Aqeel, M., Shuja, k,. H., Abbas, J., and Ziapour, A. (2021). The influence of illness perception, anxiety, and depression disorders on student mental health during COVID-19 outbreak in Pakistan: a web-based cross sectional survey. Int. J. Hum. Rights Healthc. 14, 1–20. doi: 10.1108/IJHRH-10-2020-0095

Ashraf, B. N. (2020). Stock markets' reaction to COVID-19: cases or fatalities? Res. Int. Bus. Finance 54:101249. doi: 10.1016/j.ribaf.2020.101249

Aslam, F., Mohmand, Y. T., Ferreira, P., Memon, B. A., Khan, M., and Khan, M. (2020). Network analysis of global stock markets at the beginning of the coronavirus disease (Covid-19) outbreak. Borsa Istanb. Rev. 20, 49-61. doi: 10.1016/j.bir.2020.09.003

Bagchi, B., Chatterjee, S., Ghosh, R., and Dandapat, D. (2020). Coronavirus outbreak and the great lockdown: impact on oil prices and major stock markets across the globe (Heidelberg; Singapore: Springer), 112. doi: 10.1007/978-981-15-7782-6

Baker, S. R., Bloom, N., Davis, S. J., Kost, K. J., Sammon, M. C., and Viratyosin, T. (2020). The unprecedented stock market impact of COVID-19 (No. w26945). Natl. Bur. Econ. Res . doi: 10.3386/w26945

Baldwin, R. E., and Weder, B. (2020). Economics in the Time of COVID-19 . Washington, DC: CEPR Press.

Barro, R. J., Ursua, J. F., and Weng, J. (2020). The coronavirus and the great influenza epidemic: lessons from the “Spanish Flu” for the coronavirus' potential effects on mortality and economic activity. Natl. Bur. Econ. Res . doi: 10.3386/w26866

Bash, A., and Alsaifi, K. (2019). Fear from uncertainty: an event study of Khashoggi and stock market returns. J. Behav. Exp. Finance 23, 54–58. doi: 10.1016/j.jbef.2019.05.004

Boone, L., Haugh, D., Pain, N., Salins, V., and Boone, L. (2020). Tackling the Fallout from COVID-19. Economics in the Time of COVID-19 (London: CEPR Press), 37.

Caporale, G. M., Plastun, A., and Makarenko, I. (2019). Force majeure events and stock market reactions in Ukraine. Invest. Manag. Financ. Innov. 16, 334–345. doi: 10.21511/imfi.16(1)0.2019.26

Chen, M. H., Jang, S. S., and Kim, W. G. (2007). The impact of the SARS outbreak on Taiwanese hotel stock performance: an event-study approach. Int. J. Hosp. Manag. 26, 200–212. doi: 10.1016/j.ijhm.2005.11.004

Chen, M. P., Lee, C. C., Lin, Y. H., and Chen, W. Y. (2018). Did the SARS epidemic weaken the integration of Asian stock markets? Evidence from smooth time-varying cointegration analysis. Econ. Res. Ekon. Istraz. 3, 908–926. doi: 10.1080/1331677X.2018.1456354

Chowdhury, E. K., Khan, I. I., and Dhar, B. K. (2020). Catastrophic impact of Covid-19 on the global stock markets and economic activities. Bus. Soc. Rev. 1–24. doi: 10.1111/basr.12219

Chriscaden, K. (2020). Impact of COVID-19 on People's Livelihoods, Their Health and our Food Systems [Joint statement by ILO, FAO, IFAD and WHO] . Available online at: https://www.who.int/news/item/13-10-2020-impact-of-covid-19-on-people's-livelihoods-their-health-and-our-food-systems (accessed February 13, 2021).

Chudik, A., Mohaddes, K., Pesaran, M. H., Raissi, M., and Rebucci, A. (2020). Economic Consequences of Covid-19: A Counterfactual Multi-Country analysis. VoxEU.Org . Available online at: https://voxeu.org/article/economic-consequences-covid-19-multi-country-analysis (accessed October 19, 2020).

Corbet, S., Hou, Y., Hu, Y., Lucey, B., and Oxley, L. (2020). Aye Corona! The contagion effects of being named Corona during the COVID-19 pandemic. Finance Res. Lett. 38:2020. doi: 10.2139/ssrn.3561866

Coulibaly, B., and Prasad, E. (2020). “The international monetary and financial system: how to fit it for purpose?,” in Reimagining the Global Economy: Building Back Better in a Post-COVID-19 World . Washington, DC: The Brookings Institution.

Cox, J., Greenwald, D. L., and Ludvigson, S. C. (2020). What explains the COVID-19 stock market. Natl. Bur. Econ. Res . doi: 10.3386/w27784

Cucinotta, D., and Vanelli, M. (2020). WHO declares COVID-19 a pandemic. Acta Biomed. Ateneo Parm. 9, 157–160. doi: 10.23750/abm.v91i1.9397

Czech, K., Wielechowsk, M., Kotyza, P., Benešová, I., and Laputková, A. (2020). Shaking stability: COVID-19 impact on the Visegrad group countries' financial markets. Sustain. Times 12:6282. doi: 10.3390/su12156282

Dang, T. L., and Nguyen, T. M. H. (2020). Liquidity risk and stock performance during the financial crisis. Res. Int. Bus. Finance 52:101165. doi: 10.1016/j.ribaf.2019.101165

David, S. (2020). International Business Models for a Post-COVID-19 World HLB. HLB The Global Advisory and Accounting Network . Available online at: https://www.hlb.global/international-business-models-for-a-post-covid-19-world/ (accessed February 23, 2021).

Delisle, J. (2003). SARS, greater China, and the pathologies of globalization and transition. Orbis 47, 587–604. doi: 10.1016/S0030-4387(03)00076-0

Dong, G. N., and Heo, Y. (2014). Flu epidemic, limited attention and analyst forecast behavior. Limited Attent. Analyst Forecast Behav. doi: 10.2139/ssrn.3353255

Fernandes, N. (2020). Economic effects of coronavirus outbreak (COVID-19) on the world economy . doi: 10.2139/ssrn.3557504

He, P., Sun, Y., Zhang, Y., and Li, T. (2020). COVID−19's impact on stock prices across different sectors—an event study based on the chinese stock market. Emerg. Mark. Finance Trade 56, 2198–2212. doi: 10.1080/1540496X.2020.1785865

He, Q., Liu, J., Wang, S., and Yu, J. (2020). The impact of COVID-19 on stock markets. Econ. Political Stud. 8, 275–288. doi: 10.1080/20954816.2020.1757570

Helppie McFall, B. (2011). Crash and wait? The impact of the great recession on the retirement plans of older Americans. Am. Econ. Rev. 101, 40–44. doi: 10.1257/aer.101.3.40

Hui, D. S., Azhar, D. I., Madani, T. A., Ntoumi, F., Kock, R., Dar, O., et al. (2020). The continuing 2019 nCoV epidemic threat of novel coronaviruses to global health—the latest 2019 novel coronavirus outbreak in Wuhan, China. Int. J. Infect. Dis. 91, 264–266. doi: 10.1016/j.ijid.2020.01.009

Iyke, B. N. (2020). The disease outbreak channel of exchange rate return predictability: evidence from COVID-19. Emerg. Mark. Finance Trade 56, 2277–2297. doi: 10.1080/1540496X.2020.1784718

Jabeen, S., and Rizavi, S. S. (2021). Long term and short term herding prospects: evidence from Pakistan stock exchange. Abasyn J. Manag. Sci. 14, 119–144. doi: 10.34091/AJSS.14.1.08

Jabeen, S., and Farhan, M. (2020). COVID-19: the pandemic's impact on economy and stock markets. J. Manag. Sci. 14, 29–43.

Jack, S. (2020). Covid-19: Global Stock Markets Rocket on Vaccine Hopes. BBC News . Available online at: https://www.bbc.com/news/business-54874108 (accessed November 9, 2020).

Jones, L., Palumbo, D., and Brown, D. (2021). Coronavirus: How the Pandemic has Changed the World Economy. BBC News . Available online at: https://www.bbc.com/news/business-51706225 (accessed January 24, 2021).

Kalra, R., Henderson, G. V., and Raines, G. A. (1993). Effects of the chernobyl nuclear accident on utility share prices. Q. J. Bus. Econ. 32, 52–77.

Kaplanski, G., and Levy, H. (2010). Sentiment and stock prices: the case of aviation disasters. J. Financ. Econ. 95, 174–201. doi: 10.1016/j.jfineco.2009.10.002

Karlsson, M., and Nilsson, S. (2014). The impact of the 1918 Spanish flu epidemic on economic performance in Sweden: an investigation into the consequences of an extraordinary mortality shock. J. Health Econ. 36, 1–9. doi: 10.1016/j.jhealeco.2014.03.005

Keating, J. (2001). An investigation into the cyclical incidence of dengue feve. Soc. Sci. Med. 53, 1587–1597. doi: 10.1016/S0277-9536(00)00443-3

Kharas, H., and McArthur, J. W. (2020). “Sustainable development goals: how can they be a handrail for recovery?,” in Reimagining the Global Economy: Building Back Better in a Post-COVID-19 World (Washington, DC. The Brookings Institution). Available online at: https://www.brookings.edu/multi-chapter-report/reimagining-the-global-economy-building-back-better-in-a-post-covid-19-world

Khatatbeh, I. N., Hani, M. B., and Abu-Alfoul, M. N. (2020). The impact of COVID-19 pandemic on global stock markets: an event study. Int. J. Econ. Bus. 8, 505–514. doi: 10.35808/ijeba/602

Koley, T. K., and Dhole, M. (2020). COVID-19 pandemic: The Deadly Coronavirus Outbreak in the 21st Century, 1st Edn . Oxfordshire: Routledge. doi: 10.4324/9781003095590

Lanfear, M. G., Lioui, A., and Siebert, M. G. (2018). Market anomalies and disaster risk: evidence from extreme weather events. J. Financial Mark. 46:100477. doi: 10.1016/j.finmar.2018.10.003

Lee, J. W., and McKibbin, W. J. (2004). “Estimating the global economic costs of SARS,” in Learning from SARS: Preparing for the Next Disease Outbreak , eds S. Knobler, A. Mahmoud, and S. Lemon, et al. (Washington, DC: National Academies Press).

Liu, H., Manzoor, A., Wang, C., Zhang, L., and Manzoor, Z. (2020). The COVID-19 outbreak and affected countries stock markets response. Int. J. Environ. Res. Public Health 17:2800. doi: 10.3390/ijerph17082800

Loh, E. (2006). The impact of SARS on the performance and risk profile of airline stocks. Int. J. Transp. Econ. 33, 401–422.

Lyócsa, Š., Baumöhl, E., Výrost, T., and Molná, P. (2020). Fear of the coronavirus and the stock markets. Finance Res. Lett. 36:101735. doi: 10.1016/j.frl.2020.101735

Machmuddah, Z., Utomo, S.t., Suhartono, E., Ali, S., and Ghulam, W. A. (2020). Stock market reaction to COVID-19: evidence in customer goods sector with the implication for open innovation. J. Open Innov.: Technol. Mark. Complex. 6:99. doi: 10.3390/joitmc6040099

Maqsood, A., Abbas, J., Rehman, G., and Mubeen, R. (2021). The paradigm shift for educational system continuance in the advent of COVID-19 pandemic: mental health challenges and reflections. Curr. Res. Behav. Sci. 2, 1–5. doi: 10.1016/j.crbeha.2020.100011

MckKibbin, W. J., and Sidorenko, A. A. (2006). Global macroeconomic consequences of pandemic influenza. 79.

Mctier, B. C., Tse, Y., and Wald, J. K. (2011). Do stock markets catch the flu? J. Financ. Quant. Anal. 48, 979–1000. doi: 10.1017/S0022109013000239

Mishra, M. K. (2020). The World After COVID-19 and Its Impact on Global Economy . Kiel: ZBW–Leibniz Information Centre for Economics. Available online at: https://www.econstor.eu/handle/10419/215931

Ngwakwe, C. V. (2020). Effect of COVID-19 pandemic on global stock market values: a differential analysis. Economica 16, 261–275.

Nikkinen, J., Omran, M. M., and Sahlstr, M. P. (2008). Stock returns and volatility following the september 11 attacks: evidence from 53 equity markets. Int. Rev. Financial Anal. 17, 27–46. doi: 10.1016/j.irfa.2006.12.002

Nippani, S., and Washer, K. M. (2004). SARS: a non-event for affected countries' stock markets? Appl. Financial Econ. 14, 1105–1110 doi: 10.1080/0960310042000310579

OECD (2020). OECDEmployment Outlook 2020 Facing the Jobs Crisis. Paris: OECD. Available online at: http://www.oecd.org/employment-outlook/2020 (accessed February 13, 2021).

Okorie, D. I., and Lin, B. (2021). Stock markets and the COVID-19 fractal contagion effects. Finance Res. Lett. 38:101640.

PubMed Abstract | Google Scholar

Ozili, P. K., and Arun, T. (2020). Spillover of COVID-19: impact on the global economy. doi: 10.2139/ssrn.3562570

Qin, M., Zhang, Y. C., and Su, C. W. (2020). The essential role of pandemics: a fresh insight into the oil market. Energy Res. Lett. 1, 1–6. doi: 10.46557/001c.13166

Ramelli, S., and Wagner, A. F. (2020). Feverish stock price reactions to COVID-19. Rev. Corp. Finance Stud. 9, 622–655. doi: 10.1093/rcfs/cfaa012

Ranju, P. K., and Mallikarjunappa, T. (2019). Spillover effect of MandA announcements on acquiring firms' rivals: evidence from India. Glob. Bus. Rev. 20, 692–707. doi: 10.1177/0972150919837080

Rengasamy, E. (2012). Sovereign debt crisis in the euro zone and its impact on the BRICS's stock index returns and volatility. Econ. Finance Rev. 2, 37–46. Retrieved from: https://tradingeconomics.com/pakistan/stock-market (accessed February 21, 2021).

Righi, M. B., and Ceretta, P. S. (2011). Analyzing the structural behavior of volatility in the major European markets during the Greek crisis. Econ. Bull . 31, 3016–3029.

Sansa, N. A. (2020). The impact of the COVID-19 on the financial markets: evidence from China and USA. Electron. Res. J. Soc. Sci. Human. 2:11. doi: 10.2139/ssrn.3567901

Schwert, G. W. (2011). Stock volatility during the recent financial crisis. Eur. Financ. Manag. 17, 789–805. doi: 10.1111/j.1468-036X.2011.00620.x

Shafi, M., Liu, J., and Ren, W. (2020). Impact of COVID-19 pandemic on micro, small, and medium-sized enterprises operating in Pakistan. Res. Glob. 2:100018. doi: 10.1016/j.resglo.2020.100018

Sharma, P. (2021). Coronavirus News, Markets and AI: The COVID-19 Diaries . Oxfordshire: Routledge. doi: 10.4324/9781003138976-1

Singh, B., Dhall, R., Narang, S., and Rawat, S. (2020). The outbreak of COVID-19 and stock market responses: an event study and panel data analysis for G-20 countries. Glob. Bus. Rev. 1–26. doi: 10.1177/0972150920957274

Sobieralski, J. B. (2020). Covid-19 and airline employment: insights from historical uncertainty shocks to the industry. Transp. Res. Interdiscip. Perspect. 5:100123. doi: 10.1016/j.trip.2020.100123

Su, Z., McDonnell, D., Cheshmehzangi, A., Abbas, J., Li, X., and Cai, Y. (2021). The promise and perils of Unit 731 data. BMG Glob. Health 6, 1–4. doi: 10.1136/bmjgh-2020-004772

Topcu, M., and Gulal, O. S. (2020). The impact of COVID-19 on emerging stock markets. Finance Res. Lett. 36, 101691.

UNWTO (2020). Impact Assessment of the COVID-19 Outbreak on International Tourism UNWTO . Available online at: https://www.unwto.org/impact-assessment-of-the-covid-19-outbreak-on-international-tourism (accessed February 13, 2021).

Wang, C., Wang, D., Duan, K., and Mubeen, R. (2021). Global financial crisis, smart lockdown strategies, and the COVID-19 spillover impacts: a global perspective implications from Southeast Asia. Front. Psychiatry 12:643783. doi: 10.3389/fpsyt.2021.643783

Zhang, D., Hu, M., and Ji, Q. (2020). Financial markets under the global pandemic of COVID-19. Finance Res. Lett. 36:101528. doi: 10.1016/j.frl.2020.101528

Keywords: COVID-19, stock markets, market indices, behavioral finance, SARS

Citation: Jabeen S, Farhan M, Zaka MA, Fiaz M and Farasat M (2022) COVID and World Stock Markets: A Comprehensive Discussion. Front. Psychol. 12:763346. doi: 10.3389/fpsyg.2021.763346

Received: 23 August 2021; Accepted: 30 September 2021; Published: 28 February 2022.

Reviewed by:

Copyright © 2022 Jabeen, Farhan, Zaka, Fiaz and Farasat. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Muhammad Farhan, muhammad.farhan@numl.edu.pk

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

Help | Advanced Search

Quantitative Finance > General Finance

Title: stock market prediction via deep learning techniques: a survey.

Abstract: Existing surveys on stock market prediction often focus on traditional machine learning methods instead of deep learning methods. This motivates us to provide a structured and comprehensive overview of the research on stock market prediction. We present four elaborated subtasks of stock market prediction and propose a novel taxonomy to summarize the state-of-the-art models based on deep neural networks. In addition, we also provide detailed statistics on the datasets and evaluation metrics commonly used in the stock market. Finally, we point out several future directions by sharing some new perspectives on stock market prediction.

Submission history

Access paper:.

• Other Formats

References & Citations

• Google Scholar
• Semantic Scholar

BibTeX formatted citation

Bibliographic and Citation Tools

Code, data and media associated with this article, recommenders and search tools.

• Institution

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

Stock Market Prediction Techniques: A Review Paper

• Conference paper
• First Online: 19 October 2021
• Cite this conference paper

• Kirti Sharma 18 &
• Rajni Bhalla 18  

Part of the book series: Advances in Intelligent Systems and Computing ((AISC,volume 1235))

625 Accesses

7 Citations

Due to the non-linear and highly volatile nature of the Stock market, it has become a very challenging task for researchers to make accurate predictions. Improving the efficiency of predictions has become the main goal of many researchers. From the traditional approach of working with historical dossiers to using the latest machine learning and deep learning techniques, researchers are busy finding out the best possible ways of accurate predictions. Many new models are suggested that can make good estimations of stock prices. Investors are interested in knowing both the immediate next-day prices and as well as future share prices in the long run. This paper inspects the algorithms and techniques that are useful for making accurate predictions.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

• Available as PDF
• Read on any device
• Instant download
• Own it forever
• Available as EPUB and PDF
• Compact, lightweight edition
• Dispatched in 3 to 5 business days
• Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

E. Alexander, E. Kawaler, D. Szafir, Stock Market Analysis and Prediction (2009) pp. 1–15

Google Scholar  

R. An, C. Qian, W. Zheng, W. Zheng, R. An, Stock Market Trends Prediction after Earning Release (n.d.) 06117810, 814

G. Appel, Technical Analysis: Power Tools for Active Investors . Financial Times/Prentice Hall (2005), https://books.google.co.in/books?id=RFYIAAAACAAJ

A. Atkins, M. Niranjan, E. Gerding, Financial news predicts stock market volatility better than close price. J. Finance Data Sci. 4 (2), 120–137 (2018). https://doi.org/10.1016/j.jfds.2018.02.002

Article   Google Scholar  

A. Beattie The 4 Basic Elements of Stock Value (2016). Investopedia.Com. http://www.investopedia.com/articles/fundamental-analysis/09/elements-stock-value.asp

M. Dang, D. Duong, Improvement methods for stock market prediction using financial news articles, in NICS 2016 - Proceedings of 2016 3rd National Foundation for Science and Technology Development Conference on Information and Computer Science , May , (2016), pp. 125–129. https://doi.org/10.1109/NICS.2016.7725636

O. Hegazy, O.S. Soliman, M.A. Salam. A machine learning model for stock market prediction. Int. J. Comput. Sci. Telecommun. 4 (2013) (2014). http://arxiv.org/abs/1402.7351

M. Hiransha, E.A. Gopalakrishnan, V.K. Menon, K.P. Soman, NSE stock market prediction using deep-learning models. Procedia Comput. Sci. 132 (Iccids), 1351–1362 (2018). https://doi.org/10.1016/j.procs.2018.05.050

R. Iacomin (2015). Stock market prediction, in 2015 19th International Conference on System Theory, Control and Computing, ICSTCC 2015 - Joint Conference SINTES 19, SACCS 15, SIMSIS 19 , pp. 200–205. https://doi.org/10.1109/ICSTCC.2015.7321293

B. Jeevan, E. Naresh, B.P. Vijaya Kumar, P. Kambli. Share price prediction using machine learning technique, in 2018 IEEE 3rd International Conference on Circuits, Control, Communication and Computing, I4C 2018 , 1, pp. 1–4 (2018). https://doi.org/10.1109/CIMCA.2018.8739647

S. Kalra, J.S. Prasad, Efficacy of news sentiment for stock market prediction, in International Conference on Machine Learning, Big Data, Cloud and Parallel Computing (Com-IT-Con) (2019), pp. 491–496. https://doi.org/10.1109/COMITCon.2019.8862265

W. Khan, M.A. Ghazanfar, M.A. Azam, A. Karami, K.H. Alyoubi, A.S. Alfakeeh, Stock market prediction using machine learning classifiers and social media, news. J. Ambient Intell. Hum. Comput.0123456789 (2020). https://doi.org/10.1007/s12652-020-01839-w

S. Kokila, Implementation of extended deep neural networks for stock market prediction. Int. J. Res. Appl. Sci. Eng. Technol. 8 (2), 29–34 (2020). https://doi.org/10.22214/ijraset.2020.2006

T. Mankar, T. Hotchandani, M. Madhwani, A. Chidrawar, C.S. Lifna, Stock market prediction based on social sentiments using machine learning, in 2018 International Conference on Smart City and Emerging Technology, ICSCET 2018 , (2018) pp. 1–3. https://doi.org/10.1109/ICSCET.2018.8537242

A. Nayak, M.M.M. Pai, R.M. Pai, Prediction Models for Indian Stock Market. Procedia Comput. Sci. 89 , 441–449 (2016). https://doi.org/10.1016/j.procs.2016.06.096

I.K. Nti, A.F. Adekoya, B.A. Weyori, A comprehensive evaluation of ensemble learning for stock-market prediction. J. Big Data 7 (1) (2020). https://doi.org/10.1186/s40537-020-00299-5

I.K. Nti, A.F. Adekoya, B.A. Weyori, Predicting stock market price movement using sentiment analysis: evidence from Ghana. Appl. Comput. Syst. 25 (1), 33–42 (2020). https://doi.org/10.2478/acss-2020-0004

V.S. Pagolu, K.N. Reddy, G. Panda, B. Majhi, Sentiment analysis of Twitter data for predicting stock market movements, in International Conference on Signal Processing, Communication, Power and Embedded System, SCOPES 2016 – Proceedings (2017), pp. 1345–1350. https://doi.org/10.1109/SCOPES.2016.7955659

I. Parmar, N. Agarwal, S. Saxena, R. Arora, S. Gupta, H. Dhiman, L. Chouhan, Stock Market Prediction Using Machine Learning, in ICSCCC 2018 - 1st International Conference on Secure Cyber Computing and Communications (2018), pp. 574–576. https://doi.org/10.1109/ICSCCC.2018.8703332

P.S.P. Pimpalkar, J. Karia, M. Khan, S. Anand, Mukherje, Stock market prediction using machine learning. Int. J. Adv. Eng. Res. Dev. Spec. Issue Recent Trends Data Eng. 4 (5), 6–8 (2017)

V. Rajput, S. Bobde, Stock market forecasting techniques: literature survey. Int. J. Comput. Sci. Mobile Comput. 5 (6), 500–506 (2016). www.ijcsmc.com

R.S. Deepak, S.I. Uday, D. Malathi, Machine learning approach in stock market. Int. J. Pure Appl. Math. 115 (8), 71–77 (2017)

K. Sai, R. Vanukuru, V. Kranthi, S. Reddy, Stock market prediction using machine learning stock market prediction using machine learning view project development of microcontroller based automated public transport ticketing system view project stock market prediction using machine learning. Int. Res. J. Eng. Technol. 1032 (2008). https://doi.org/10.13140/RG.2.2.12300.77448

V.K. Sai Redddy, Stock market prediction using machine learning. Int. Res. J. Eng. Technol. 05 (10) (2018). https://doi.org/10.1109/ICAC49085.2019.9103381

D. Selvamuthu, V. Kumar, A. Mishra, Indian stock market prediction using artificial neural networks on tick data. Financ. Innovation 5 (1) (2019). https://doi.org/10.1186/s40854-019-0131-7

D.S. Shanthi, T. Aarthi, A.K. Bhuvanesh, R.A.C. Prabha, Pattern Recogn. Stock Market 9 (3), 106–111 (2020)

S. Sharma, B. Kaushik, Quantitative analysis of stock market prediction for accurate investment decisions in future. J. Artif. Intell. 11 (1), 48–54 (2018). https://doi.org/10.3923/jai.2018.48.54

H.J. Shunrong Shen, Tongda Zhang, Stock market forecasting using machine learning algorithms. Successful User Experience: Strategies and Roadmaps, 1–344 (2015). https://doi.org/10.1016/c2013-0-19353-1

J. Si, A. Mukherjee, B. Liu, Q. Li, H. Li, X. Deng, Exploiting topic based twitter sentiment for stock prediction, in ACL 2013 - 51st Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference , 2 (2011) (2013), pp. 24–29

S. Singh, T.K. Madan, J. Kumar, A.K. Singh, Stock market forecasting using machine learning: today and tomorrow, in 2nd International Conference on Intelligent Computing, Instrumentation and Control Technologies (ICICICT) (2019). https://doi.org/10.1109/ASYU48272.2019.8946372

Understanding Fundamental and Technical Analysis. (n.d.). Retrieved August 29 2020, from https://groww.in/blog/fundamental-vs-technical-analysis/#2Factors_important_for_the_analysis

T. Xia, Q. Sun, A. Zhou, S. Wang, S. Xiong, S. Gao, J. Li, Q. Yuan, Improving the performance of stock trend prediction by applying GA to feature selection, in Proceedings - 8th IEEE International Symposium on Cloud and Services Computing, SC2 2018 , (2018), pp. 122–126. https://doi.org/10.1109/SC2.2018.00025

X. Zhou, Z. Pan, G. Hu, S. Tang, C. Zhao, Stock market prediction on high-frequency data using generative adversarial nets. Math. Prob. Eng. 2018 (2018)

Download references

Author information

Authors and affiliations.

School of Computer Application, Lovely Professional University, Jalandhar, India

Kirti Sharma & Rajni Bhalla

You can also search for this author in PubMed   Google Scholar

Editor information

Editors and affiliations.

The PNG University of Technology, Lae, Papua New Guinea

Ashish Kumar Luhach

CHRIST (Deemed to be University), Bangalore, India

Ramesh Chandra Poonia

University of Eastern Finland, Kuopio, Finland

Xiao-Zhi Gao

Namibia University of Science and Technology, Windhoek, Namibia

Dharm Singh Jat

Rights and permissions

Reprints and permissions

Copyright information

© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

About this paper

Cite this paper.

Sharma, K., Bhalla, R. (2022). Stock Market Prediction Techniques: A Review Paper. In: Luhach, A.K., Poonia, R.C., Gao, XZ., Singh Jat, D. (eds) Second International Conference on Sustainable Technologies for Computational Intelligence. Advances in Intelligent Systems and Computing, vol 1235. Springer, Singapore. https://doi.org/10.1007/978-981-16-4641-6_15

Download citation

DOI : https://doi.org/10.1007/978-981-16-4641-6_15

Published : 19 October 2021

Publisher Name : Springer, Singapore

Print ISBN : 978-981-16-4640-9

Online ISBN : 978-981-16-4641-6

eBook Packages : Intelligent Technologies and Robotics Intelligent Technologies and Robotics (R0)

Share this paper

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Publish with us

Policies and ethics

• Find a journal
• Track your research

• Open access
• Published: 28 August 2020

Short-term stock market price trend prediction using a comprehensive deep learning system

• Jingyi Shen 1 &
• M. Omair Shafiq   ORCID: orcid.org/0000-0002-1859-8296 1  

Journal of Big Data volume  7 , Article number:  66 ( 2020 ) Cite this article

268k Accesses

164 Citations

91 Altmetric

Metrics details

In the era of big data, deep learning for predicting stock market prices and trends has become even more popular than before. We collected 2 years of data from Chinese stock market and proposed a comprehensive customization of feature engineering and deep learning-based model for predicting price trend of stock markets. The proposed solution is comprehensive as it includes pre-processing of the stock market dataset, utilization of multiple feature engineering techniques, combined with a customized deep learning based system for stock market price trend prediction. We conducted comprehensive evaluations on frequently used machine learning models and conclude that our proposed solution outperforms due to the comprehensive feature engineering that we built. The system achieves overall high accuracy for stock market trend prediction. With the detailed design and evaluation of prediction term lengths, feature engineering, and data pre-processing methods, this work contributes to the stock analysis research community both in the financial and technical domains.

Introduction

Stock market is one of the major fields that investors are dedicated to, thus stock market price trend prediction is always a hot topic for researchers from both financial and technical domains. In this research, our objective is to build a state-of-art prediction model for price trend prediction, which focuses on short-term price trend prediction.

As concluded by Fama in [ 26 ], financial time series prediction is known to be a notoriously difficult task due to the generally accepted, semi-strong form of market efficiency and the high level of noise. Back in 2003, Wang et al. in [ 44 ] already applied artificial neural networks on stock market price prediction and focused on volume, as a specific feature of stock market. One of the key findings by them was that the volume was not found to be effective in improving the forecasting performance on the datasets they used, which was S&P 500 and DJI. Ince and Trafalis in [ 15 ] targeted short-term forecasting and applied support vector machine (SVM) model on the stock price prediction. Their main contribution is performing a comparison between multi-layer perceptron (MLP) and SVM then found that most of the scenarios SVM outperformed MLP, while the result was also affected by different trading strategies. In the meantime, researchers from financial domains were applying conventional statistical methods and signal processing techniques on analyzing stock market data.

The optimization techniques, such as principal component analysis (PCA) were also applied in short-term stock price prediction [ 22 ]. During the years, researchers are not only focused on stock price-related analysis but also tried to analyze stock market transactions such as volume burst risks, which expands the stock market analysis research domain broader and indicates this research domain still has high potential [ 39 ]. As the artificial intelligence techniques evolved in recent years, many proposed solutions attempted to combine machine learning and deep learning techniques based on previous approaches, and then proposed new metrics that serve as training features such as Liu and Wang [ 23 ]. This type of previous works belongs to the feature engineering domain and can be considered as the inspiration of feature extension ideas in our research. Liu et al. in [ 24 ] proposed a convolutional neural network (CNN) as well as a long short-term memory (LSTM) neural network based model to analyze different quantitative strategies in stock markets. The CNN serves for the stock selection strategy, automatically extracts features based on quantitative data, then follows an LSTM to preserve the time-series features for improving profits.

The latest work also proposes a similar hybrid neural network architecture, integrating a convolutional neural network with a bidirectional long short-term memory to predict the stock market index [ 4 ]. While the researchers frequently proposed different neural network solution architectures, it brought further discussions about the topic if the high cost of training such models is worth the result or not.

There are three key contributions of our work (1) a new dataset extracted and cleansed (2) a comprehensive feature engineering, and (3) a customized long short-term memory (LSTM) based deep learning model.

We have built the dataset by ourselves from the data source as an open-sourced data API called Tushare [ 43 ]. The novelty of our proposed solution is that we proposed a feature engineering along with a fine-tuned system instead of just an LSTM model only. We observe from the previous works and find the gaps and proposed a solution architecture with a comprehensive feature engineering procedure before training the prediction model. With the success of feature extension method collaborating with recursive feature elimination algorithms, it opens doors for many other machine learning algorithms to achieve high accuracy scores for short-term price trend prediction. It proved the effectiveness of our proposed feature extension as feature engineering. We further introduced our customized LSTM model and further improved the prediction scores in all the evaluation metrics. The proposed solution outperformed the machine learning and deep learning-based models in similar previous works.

The remainder of this paper is organized as follows. “ Survey of related works ” section describes the survey of related works. “ The dataset ” section provides details on the data that we extracted from the public data sources and the dataset prepared. “ Methods ” section presents the research problems, methods, and design of the proposed solution. Detailed technical design with algorithms and how the model implemented are also included in this section. “ Results ” section presents comprehensive results and evaluation of our proposed model, and by comparing it with the models used in most of the related works. “ Discussion ” section provides a discussion and comparison of the results. “ Conclusion ” section presents the conclusion. This research paper has been built based on Shen [ 36 ].

Survey of related works

In this section, we discuss related works. We reviewed the related work in two different domains: technical and financial, respectively.

Kim and Han in [ 19 ] built a model as a combination of artificial neural networks (ANN) and genetic algorithms (GAs) with discretization of features for predicting stock price index. The data used in their study include the technical indicators as well as the direction of change in the daily Korea stock price index (KOSPI). They used the data containing samples of 2928 trading days, ranging from January 1989 to December 1998, and give their selected features and formulas. They also applied optimization of feature discretization, as a technique that is similar to dimensionality reduction. The strengths of their work are that they introduced GA to optimize the ANN. First, the amount of input features and processing elements in the hidden layer are 12 and not adjustable. Another limitation is in the learning process of ANN, and the authors only focused on two factors in optimization. While they still believed that GA has great potential for feature discretization optimization. Our initialized feature pool refers to the selected features. Qiu and Song in [ 34 ] also presented a solution to predict the direction of the Japanese stock market based on an optimized artificial neural network model. In this work, authors utilize genetic algorithms together with artificial neural network based models, and name it as a hybrid GA-ANN model.

Piramuthu in [ 33 ] conducted a thorough evaluation of different feature selection methods for data mining applications. He used for datasets, which were credit approval data, loan defaults data, web traffic data, tam, and kiang data, and compared how different feature selection methods optimized decision tree performance. The feature selection methods he compared included probabilistic distance measure: the Bhattacharyya measure, the Matusita measure, the divergence measure, the Mahalanobis distance measure, and the Patrick-Fisher measure. For inter-class distance measures: the Minkowski distance measure, city block distance measure, Euclidean distance measure, the Chebychev distance measure, and the nonlinear (Parzen and hyper-spherical kernel) distance measure. The strength of this paper is that the author evaluated both probabilistic distance-based and several inter-class feature selection methods. Besides, the author performed the evaluation based on different datasets, which reinforced the strength of this paper. However, the evaluation algorithm was a decision tree only. We cannot conclude if the feature selection methods will still perform the same on a larger dataset or a more complex model.

Hassan and Nath in [ 9 ] applied the Hidden Markov Model (HMM) on the stock market forecasting on stock prices of four different Airlines. They reduce states of the model into four states: the opening price, closing price, the highest price, and the lowest price. The strong point of this paper is that the approach does not need expert knowledge to build a prediction model. While this work is limited within the industry of Airlines and evaluated on a very small dataset, it may not lead to a prediction model with generality. One of the approaches in stock market prediction related works could be exploited to do the comparison work. The authors selected a maximum 2 years as the date range of training and testing dataset, which provided us a date range reference for our evaluation part.

Lei in [ 21 ] exploited Wavelet Neural Network (WNN) to predict stock price trends. The author also applied Rough Set (RS) for attribute reduction as an optimization. Rough Set was utilized to reduce the stock price trend feature dimensions. It was also used to determine the structure of the Wavelet Neural Network. The dataset of this work consists of five well-known stock market indices, i.e., (1) SSE Composite Index (China), (2) CSI 300 Index (China), (3) All Ordinaries Index (Australian), (4) Nikkei 225 Index (Japan), and (5) Dow Jones Index (USA). Evaluation of the model was based on different stock market indices, and the result was convincing with generality. By using Rough Set for optimizing the feature dimension before processing reduces the computational complexity. However, the author only stressed the parameter adjustment in the discussion part but did not specify the weakness of the model itself. Meanwhile, we also found that the evaluations were performed on indices, the same model may not have the same performance if applied on a specific stock.

Lee in [ 20 ] used the support vector machine (SVM) along with a hybrid feature selection method to carry out prediction of stock trends. The dataset in this research is a sub dataset of NASDAQ Index in Taiwan Economic Journal Database (TEJD) in 2008. The feature selection part was using a hybrid method, supported sequential forward search (SSFS) played the role of the wrapper. Another advantage of this work is that they designed a detailed procedure of parameter adjustment with performance under different parameter values. The clear structure of the feature selection model is also heuristic to the primary stage of model structuring. One of the limitations was that the performance of SVM was compared to back-propagation neural network (BPNN) only and did not compare to the other machine learning algorithms.

Sirignano and Cont leveraged a deep learning solution trained on a universal feature set of financial markets in [ 40 ]. The dataset used included buy and sell records of all transactions, and cancellations of orders for approximately 1000 NASDAQ stocks through the order book of the stock exchange. The NN consists of three layers with LSTM units and a feed-forward layer with rectified linear units (ReLUs) at last, with stochastic gradient descent (SGD) algorithm as an optimization. Their universal model was able to generalize and cover the stocks other than the ones in the training data. Though they mentioned the advantages of a universal model, the training cost was still expensive. Meanwhile, due to the inexplicit programming of the deep learning algorithm, it is unclear that if there are useless features contaminated when feeding the data into the model. Authors found out that it would have been better if they performed feature selection part before training the model and found it as an effective way to reduce the computational complexity.

Ni et al. in [ 30 ] predicted stock price trends by exploiting SVM and performed fractal feature selection for optimization. The dataset they used is the Shanghai Stock Exchange Composite Index (SSECI), with 19 technical indicators as features. Before processing the data, they optimized the input data by performing feature selection. When finding the best parameter combination, they also used a grid search method, which is k cross-validation. Besides, the evaluation of different feature selection methods is also comprehensive. As the authors mentioned in their conclusion part, they only considered the technical indicators but not macro and micro factors in the financial domain. The source of datasets that the authors used was similar to our dataset, which makes their evaluation results useful to our research. They also mentioned a method called k cross-validation when testing hyper-parameter combinations.

McNally et al. in [ 27 ] leveraged RNN and LSTM on predicting the price of Bitcoin, optimized by using the Boruta algorithm for feature engineering part, and it works similarly to the random forest classifier. Besides feature selection, they also used Bayesian optimization to select LSTM parameters. The Bitcoin dataset ranged from the 19th of August 2013 to 19th of July 2016. Used multiple optimization methods to improve the performance of deep learning methods. The primary problem of their work is overfitting. The research problem of predicting Bitcoin price trend has some similarities with stock market price prediction. Hidden features and noises embedded in the price data are threats of this work. The authors treated the research question as a time sequence problem. The best part of this paper is the feature engineering and optimization part; we could replicate the methods they exploited in our data pre-processing.

Weng et al. in [ 45 ] focused on short-term stock price prediction by using ensemble methods of four well-known machine learning models. The dataset for this research is five sets of data. They obtained these datasets from three open-sourced APIs and an R package named TTR. The machine learning models they used are (1) neural network regression ensemble (NNRE), (2) a Random Forest with unpruned regression trees as base learners (RFR), (3) AdaBoost with unpruned regression trees as base learners (BRT) and (4) a support vector regression ensemble (SVRE). A thorough study of ensemble methods specified for short-term stock price prediction. With background knowledge, the authors selected eight technical indicators in this study then performed a thoughtful evaluation of five datasets. The primary contribution of this paper is that they developed a platform for investors using R, which does not need users to input their own data but call API to fetch the data from online source straightforward. From the research perspective, they only evaluated the prediction of the price for 1 up to 10 days ahead but did not evaluate longer terms than two trading weeks or a shorter term than 1 day. The primary limitation of their research was that they only analyzed 20 U.S.-based stocks, the model might not be generalized to other stock market or need further revalidation to see if it suffered from overfitting problems.

Kara et al. in [ 17 ] also exploited ANN and SVM in predicting the movement of stock price index. The data set they used covers a time period from January 2, 1997, to December 31, 2007, of the Istanbul Stock Exchange. The primary strength of this work is its detailed record of parameter adjustment procedures. While the weaknesses of this work are that neither the technical indicator nor the model structure has novelty, and the authors did not explain how their model performed better than other models in previous works. Thus, more validation works on other datasets would help. They explained how ANN and SVM work with stock market features, also recorded the parameter adjustment. The implementation part of our research could benefit from this previous work.

Jeon et al. in [ 16 ] performed research on millisecond interval-based big dataset by using pattern graph tracking to complete stock price prediction tasks. The dataset they used is a millisecond interval-based big dataset of historical stock data from KOSCOM, from August 2014 to October 2014, 10G–15G capacity. The author applied Euclidean distance, Dynamic Time Warping (DTW) for pattern recognition. For feature selection, they used stepwise regression. The authors completed the prediction task by ANN and Hadoop and RHive for big data processing. The “ Results ” section is based on the result processed by a combination of SAX and Jaro–Winkler distance. Before processing the data, they generated aggregated data at 5-min intervals from discrete data. The primary strength of this work is the explicit structure of the whole implementation procedure. While they exploited a relatively old model, another weakness is the overall time span of the training dataset is extremely short. It is difficult to access the millisecond interval-based data in real life, so the model is not as practical as a daily based data model.

Huang et al. in [ 12 ] applied a fuzzy-GA model to complete the stock selection task. They used the key stocks of the 200 largest market capitalization listed as the investment universe in the Taiwan Stock Exchange. Besides, the yearly financial statement data and the stock returns were taken from the Taiwan Economic Journal (TEJ) database at www.tej.com.tw/ for the time period from year 1995 to year 2009. They conducted the fuzzy membership function with model parameters optimized with GA and extracted features for optimizing stock scoring. The authors proposed an optimized model for selection and scoring of stocks. Different from the prediction model, the authors more focused on stock rankings, selection, and performance evaluation. Their structure is more practical among investors. But in the model validation part, they did not compare the model with existed algorithms but the statistics of the benchmark, which made it challenging to identify if GA would outperform other algorithms.

Fischer and Krauss in [ 5 ] applied long short-term memory (LSTM) on financial market prediction. The dataset they used is S&P 500 index constituents from Thomson Reuters. They obtained all month-end constituent lists for the S&P 500 from Dec 1989 to Sep 2015, then consolidated the lists into a binary matrix to eliminate survivor bias. The authors also used RMSprop as an optimizer, which is a mini-batch version of rprop. The primary strength of this work is that the authors used the latest deep learning technique to perform predictions. They relied on the LSTM technique, lack of background knowledge in the financial domain. Although the LSTM outperformed the standard DNN and logistic regression algorithms, while the author did not mention the effort to train an LSTM with long-time dependencies.

Tsai and Hsiao in [ 42 ] proposed a solution as a combination of different feature selection methods for prediction of stocks. They used Taiwan Economic Journal (TEJ) database as data source. The data used in their analysis was from year 2000 to 2007. In their work, they used a sliding window method and combined it with multi layer perceptron (MLP) based artificial neural networks with back propagation, as their prediction model. In their work, they also applied principal component analysis (PCA) for dimensionality reduction, genetic algorithms (GA) and the classification and regression trees (CART) to select important features. They did not just rely on technical indices only. Instead, they also included both fundamental and macroeconomic indices in their analysis. The authors also reported a comparison on feature selection methods. The validation part was done by combining the model performance stats with statistical analysis.

Pimenta et al. in [ 32 ] leveraged an automated investing method by using multi-objective genetic programming and applied it in the stock market. The dataset was obtained from Brazilian stock exchange market (BOVESPA), and the primary techniques they exploited were a combination of multi-objective optimization, genetic programming, and technical trading rules. For optimization, they leveraged genetic programming (GP) to optimize decision rules. The novelty of this paper was in the evaluation part. They included a historical period, which was a critical moment of Brazilian politics and economics when performing validation. This approach reinforced the generalization strength of their proposed model. When selecting the sub-dataset for evaluation, they also set criteria to ensure more asset liquidity. While the baseline of the comparison was too basic and fundamental, and the authors did not perform any comparison with other existing models.

Huang and Tsai in [ 13 ] conducted a filter-based feature selection assembled with a hybrid self-organizing feature map (SOFM) support vector regression (SVR) model to forecast Taiwan index futures (FITX) trend. They divided the training samples into clusters to marginally improve the training efficiency. The authors proposed a comprehensive model, which was a combination of two novel machine learning techniques in stock market analysis. Besides, the optimizer of feature selection was also applied before the data processing to improve the prediction accuracy and reduce the computational complexity of processing daily stock index data. Though they optimized the feature selection part and split the sample data into small clusters, it was already strenuous to train daily stock index data of this model. It would be difficult for this model to predict trading activities in shorter time intervals since the data volume would be increased drastically. Moreover, the evaluation is not strong enough since they set a single SVR model as a baseline, but did not compare the performance with other previous works, which caused difficulty for future researchers to identify the advantages of SOFM-SVR model why it outperforms other algorithms.

Thakur and Kumar in [ 41 ] also developed a hybrid financial trading support system by exploiting multi-category classifiers and random forest (RAF). They conducted their research on stock indices from NASDAQ, DOW JONES, S&P 500, NIFTY 50, and NIFTY BANK. The authors proposed a hybrid model combined random forest (RF) algorithms with a weighted multicategory generalized eigenvalue support vector machine (WMGEPSVM) to generate “Buy/Hold/Sell” signals. Before processing the data, they used Random Forest (RF) for feature pruning. The authors proposed a practical model designed for real-life investment activities, which could generate three basic signals for investors to refer to. They also performed a thorough comparison of related algorithms. While they did not mention the time and computational complexity of their works. Meanwhile, the unignorable issue of their work was the lack of financial domain knowledge background. The investors regard the indices data as one of the attributes but could not take the signal from indices to operate a specific stock straightforward.

Hsu in [ 11 ] assembled feature selection with a back propagation neural network (BNN) combined with genetic programming to predict the stock/futures price. The dataset in this research was obtained from Taiwan Stock Exchange Corporation (TWSE). The authors have introduced the description of the background knowledge in detail. While the weakness of their work is that it is a lack of data set description. This is a combination of the model proposed by other previous works. Though we did not see the novelty of this work, we can still conclude that the genetic programming (GP) algorithm is admitted in stock market research domain. To reinforce the validation strengths, it would be good to consider adding GP models into evaluation if the model is predicting a specific price.

Hafezi et al. in [ 7 ] built a bat-neural network multi-agent system (BN-NMAS) to predict stock price. The dataset was obtained from the Deutsche bundes-bank. They also applied the Bat algorithm (BA) for optimizing neural network weights. The authors illustrated their overall structure and logic of system design in clear flowcharts. While there were very few previous works that had performed on DAX data, it would be difficult to recognize if the model they proposed still has the generality if migrated on other datasets. The system design and feature selection logic are fascinating, which worth referring to. Their findings in optimization algorithms are also valuable for the research in the stock market price prediction research domain. It is worth trying the Bat algorithm (BA) when constructing neural network models.

Long et al. in [ 25 ] conducted a deep learning approach to predict the stock price movement. The dataset they used is the Chinese stock market index CSI 300. For predicting the stock price movement, they constructed a multi-filter neural network (MFNN) with stochastic gradient descent (SGD) and back propagation optimizer for learning NN parameters. The strength of this paper is that the authors exploited a novel model with a hybrid model constructed by different kinds of neural networks, it provides an inspiration for constructing hybrid neural network structures.

Atsalakis and Valavanis in [ 1 ] proposed a solution of a neuro-fuzzy system, which is composed of controller named as Adaptive Neuro Fuzzy Inference System (ANFIS), to achieve short-term stock price trend prediction. The noticeable strength of this work is the evaluation part. Not only did they compare their proposed system with the popular data models, but also compared with investment strategies. While the weakness that we found from their proposed solution is that their solution architecture is lack of optimization part, which might limit their model performance. Since our proposed solution is also focusing on short-term stock price trend prediction, this work is heuristic for our system design. Meanwhile, by comparing with the popular trading strategies from investors, their work inspired us to compare the strategies used by investors with techniques used by researchers.

Nekoeiqachkanloo et al. in [ 29 ] proposed a system with two different approaches for stock investment. The strengths of their proposed solution are obvious. First, it is a comprehensive system that consists of data pre-processing and two different algorithms to suggest the best investment portions. Second, the system also embedded with a forecasting component, which also retains the features of the time series. Last but not least, their input features are a mix of fundamental features and technical indices that aim to fill in the gap between the financial domain and technical domain. However, their work has a weakness in the evaluation part. Instead of evaluating the proposed system on a large dataset, they chose 25 well-known stocks. There is a high possibility that the well-known stocks might potentially share some common hidden features.

As another related latest work, Idrees et al. [ 14 ] published a time series-based prediction approach for the volatility of the stock market. ARIMA is not a new approach in the time series prediction research domain. Their work is more focusing on the feature engineering side. Before feeding the features into ARIMA models, they designed three steps for feature engineering: Analyze the time series, identify if the time series is stationary or not, perform estimation by plot ACF and PACF charts and look for parameters. The only weakness of their proposed solution is that the authors did not perform any customization on the existing ARIMA model, which might limit the system performance to be improved.

One of the main weaknesses found in the related works is limited data-preprocessing mechanisms built and used. Technical works mostly tend to focus on building prediction models. When they select the features, they list all the features mentioned in previous works and go through the feature selection algorithm then select the best-voted features. Related works in the investment domain have shown more interest in behavior analysis, such as how herding behaviors affect the stock performance, or how the percentage of inside directors hold the firm’s common stock affects the performance of a certain stock. These behaviors often need a pre-processing procedure of standard technical indices and investment experience to recognize.

In the related works, often a thorough statistical analysis is performed based on a special dataset and conclude new features rather than performing feature selections. Some data, such as the percentage of a certain index fluctuation has been proven to be effective on stock performance. We believe that by extracting new features from data, then combining such features with existed common technical indices will significantly benefit the existing and well-tested prediction models.

The dataset

This section details the data that was extracted from the public data sources, and the final dataset that was prepared. Stock market-related data are diverse, so we first compared the related works from the survey of financial research works in stock market data analysis to specify the data collection directions. After collecting the data, we defined a data structure of the dataset. Given below, we describe the dataset in detail, including the data structure, and data tables in each category of data with the segment definitions.

Description of our dataset

In this section, we will describe the dataset in detail. This dataset consists of 3558 stocks from the Chinese stock market. Besides the daily price data, daily fundamental data of each stock ID, we also collected the suspending and resuming history, top 10 shareholders, etc. We list two reasons that we choose 2 years as the time span of this dataset: (1) most of the investors perform stock market price trend analysis using the data within the latest 2 years, (2) using more recent data would benefit the analysis result. We collected data through the open-sourced API, namely Tushare [ 43 ], mean-while we also leveraged a web-scraping technique to collect data from Sina Finance web pages, SWS Research website.

Data structure

Figure  1 illustrates all the data tables in the dataset. We collected four categories of data in this dataset: (1) basic data, (2) trading data, (3) finance data, and (4) other reference data. All the data tables can be linked to each other by a common field called “Stock ID” It is a unique stock identifier registered in the Chinese Stock market. Table  1 shows an overview of the dataset.

Data structure for the extracted dataset

The Table  1 lists the field information of each data table as well as which category the data table belongs to.

In this section, we present the proposed methods and the design of the proposed solution. Moreover, we also introduce the architecture design as well as algorithmic and implementation details.

Problem statement

We analyzed the best possible approach for predicting short-term price trends from different aspects: feature engineering, financial domain knowledge, and prediction algorithm. Then we addressed three research questions in each aspect, respectively: How can feature engineering benefit model prediction accuracy? How do findings from the financial domain benefit prediction model design? And what is the best algorithm for predicting short-term price trends?

The first research question is about feature engineering. We would like to know how the feature selection method benefits the performance of prediction models. From the abundance of the previous works, we can conclude that stock price data embedded with a high level of noise, and there are also correlations between features, which makes the price prediction notoriously difficult. That is also the primary reason for most of the previous works introduced the feature engineering part as an optimization module.

The second research question is evaluating the effectiveness of findings we extracted from the financial domain. Different from the previous works, besides the common evaluation of data models such as the training costs and scores, our evaluation will emphasize the effectiveness of newly added features that we extracted from the financial domain. We introduce some features from the financial domain. While we only obtained some specific findings from previous works, and the related raw data needs to be processed into usable features. After extracting related features from the financial domain, we combine the features with other common technical indices for voting out the features with a higher impact. There are numerous features said to be effective from the financial domain, and it would be impossible for us to cover all of them. Thus, how to appropriately convert the findings from the financial domain to a data processing module of our system design is a hidden research question that we attempt to answer.

The third research question is that which algorithms are we going to model our data? From the previous works, researchers have been putting efforts into the exact price prediction. We decompose the problem into predicting the trend and then the exact number. This paper focuses on the first step. Hence, the objective has been converted to resolve a binary classification problem, meanwhile, finding an effective way to eliminate the negative effect brought by the high level of noise. Our approach is to decompose the complex problem into sub-problems which have fewer dependencies and resolve them one by one, and then compile the resolutions into an ensemble model as an aiding system for investing behavior reference.

In the previous works, researchers have been using a variety of models for predicting stock price trends. While most of the best-performed models are based on machine learning techniques, in this work, we will compare our approach with the outperformed machine learning models in the evaluation part and find the solution for this research question.

Proposed solution

The high-level architecture of our proposed solution could be separated into three parts. First is the feature selection part, to guarantee the selected features are highly effective. Second, we look into the data and perform the dimensionality reduction. And the last part, which is the main contribution of our work is to build a prediction model of target stocks. Figure  2 depicts a high-level architecture of the proposed solution.

High-level architecture of the proposed solution

There are ways to classify different categories of stocks. Some investors prefer long-term investments, while others show more interest in short-term investments. It is common to see the stock-related reports showing an average performance, while the stock price is increasing drastically; this is one of the phenomena that indicate the stock price prediction has no fixed rules, thus finding effective features before training a model on data is necessary.

In this research, we focus on the short-term price trend prediction. Currently, we only have the raw data with no labels. So, the very first step is to label the data. We mark the price trend by comparing the current closing price with the closing price of n trading days ago, the range of n is from 1 to 10 since our research is focusing on the short-term. If the price trend goes up, we mark it as 1 or mark as 0 in the opposite case. To be more specified, we use the indices from the indices of n  −  1 th day to predict the price trend of the n th day.

According to the previous works, some researchers who applied both financial domain knowledge and technical methods on stock data were using rules to filter the high-quality stocks. We referred to their works and exploited their rules to contribute to our feature extension design.

However, to ensure the best performance of the prediction model, we will look into the data first. There are a large number of features in the raw data; if we involve all the features into our consideration, it will not only drastically increase the computational complexity but will also cause side effects if we would like to perform unsupervised learning in further research. So, we leverage the recursive feature elimination (RFE) to ensure all the selected features are effective.

We found most of the previous works in the technical domain were analyzing all the stocks, while in the financial domain, researchers prefer to analyze the specific scenario of investment, to fill the gap between the two domains, we decide to apply a feature extension based on the findings we gathered from the financial domain before we start the RFE procedure.

Since we plan to model the data into time series, the number of the features, the more complex the training procedure will be. So, we will leverage the dimensionality reduction by using randomized PCA at the beginning of our proposed solution architecture.

Detailed technical design elaboration

This section provides an elaboration of the detailed technical design as being a comprehensive solution based on utilizing, combining, and customizing several existing data preprocessing, feature engineering, and deep learning techniques. Figure  3 provides the detailed technical design from data processing to prediction, including the data exploration. We split the content by main procedures, and each procedure contains algorithmic steps. Algorithmic details are elaborated in the next section. The contents of this section will focus on illustrating the data workflow.

Detailed technical design of the proposed solution

Based on the literature review, we select the most commonly used technical indices and then feed them into the feature extension procedure to get the expanded feature set. We will select the most effective i features from the expanded feature set. Then we will feed the data with i selected features into the PCA algorithm to reduce the dimension into j features. After we get the best combination of i and j , we process the data into finalized the feature set and feed them into the LSTM [ 10 ] model to get the price trend prediction result.

The novelty of our proposed solution is that we will not only apply the technical method on raw data but also carry out the feature extensions that are used among stock market investors. Details on feature extension are given in the next subsection. Experiences gained from applying and optimizing deep learning based solutions in [ 37 , 38 ] were taken into account while designing and customizing feature engineering and deep learning solution in this work.

Applying feature extension

The first main procedure in Fig.  3 is the feature extension. In this block, the input data is the most commonly used technical indices concluded from related works. The three feature extension methods are max–min scaling, polarizing, and calculating fluctuation percentage. Not all the technical indices are applicable for all three of the feature extension methods; this procedure only applies the meaningful extension methods on technical indices. We choose meaningful extension methods while looking at how the indices are calculated. The technical indices and the corresponding feature extension methods are illustrated in Table  2 .

After the feature extension procedure, the expanded features will be combined with the most commonly used technical indices, i.e., input data with output data, and feed into RFE block as input data in the next step.

Applying recursive feature elimination

After the feature extension above, we explore the most effective i features by using the Recursive Feature Elimination (RFE) algorithm [ 6 ]. We estimate all the features by two attributes, coefficient, and feature importance. We also limit the features that remove from the pool by one, which means we will remove one feature at each step and retain all the relevant features. Then the output of the RFE block will be the input of the next step, which refers to PCA.

Applying principal component analysis (PCA)

The very first step before leveraging PCA is feature pre-processing. Because some of the features after RFE are percentage data, while others are very large numbers, i.e., the output from RFE are in different units. It will affect the principal component extraction result. Thus, before feeding the data into the PCA algorithm [ 8 ], a feature pre-processing is necessary. We also illustrate the effectiveness and methods comparison in “ Results ” section.

After performing feature pre-processing, the next step is to feed the processed data with selected i features into the PCA algorithm to reduce the feature matrix scale into j features. This step is to retain as many effective features as possible and meanwhile eliminate the computational complexity of training the model. This research work also evaluates the best combination of i and j, which has relatively better prediction accuracy, meanwhile, cuts the computational consumption. The result can be found in the “ Results ” section, as well. After the PCA step, the system will get a reshaped matrix with j columns.

Fitting long short-term memory (LSTM) model

PCA reduced the dimensions of the input data, while the data pre-processing is mandatory before feeding the data into the LSTM layer. The reason for adding the data pre-processing step before the LSTM model is that the input matrix formed by principal components has no time steps. While one of the most important parameters of training an LSTM is the number of time steps. Hence, we have to model the matrix into corresponding time steps for both training and testing dataset.

After performing the data pre-processing part, the last step is to feed the training data into LSTM and evaluate the performance using testing data. As a variant neural network of RNN, even with one LSTM layer, the NN structure is still a deep neural network since it can process sequential data and memorizes its hidden states through time. An LSTM layer is composed of one or more LSTM units, and an LSTM unit consists of cells and gates to perform classification and prediction based on time series data.

The LSTM structure is formed by two layers. The input dimension is determined by j after the PCA algorithm. The first layer is the input LSTM layer, and the second layer is the output layer. The final output will be 0 or 1 indicates if the stock price trend prediction result is going down or going up, as a supporting suggestion for the investors to perform the next investment decision.

Design discussion

Feature extension is one of the novelties of our proposed price trend predicting system. In the feature extension procedure, we use technical indices to collaborate with the heuristic processing methods learned from investors, which fills the gap between the financial research area and technical research area.

Since we proposed a system of price trend prediction, feature engineering is extremely important to the final prediction result. Not only the feature extension method is helpful to guarantee we do not miss the potentially correlated feature, but also feature selection method is necessary for pooling the effective features. The more irrelevant features are fed into the model, the more noise would be introduced. Each main procedure is carefully considered contributing to the whole system design.

Besides the feature engineering part, we also leverage LSTM, the state-of-the-art deep learning method for time-series prediction, which guarantees the prediction model can capture both complex hidden pattern and the time-series related pattern.

It is known that the training cost of deep learning models is expansive in both time and hardware aspects; another advantage of our system design is the optimization procedure—PCA. It can retain the principal components of the features while reducing the scale of the feature matrix, thus help the system to save the training cost of processing the large time-series feature matrix.

Algorithm elaboration

This section provides comprehensive details on the algorithms we built while utilizing and customizing different existing techniques. Details about the terminologies, parameters, as well as optimizers. From the legend on the right side of Fig.  3 , we note the algorithm steps as octagons, all of them can be found in this “ Algorithm elaboration ” section.

Before dive deep into the algorithm steps, here is the brief introduction of data pre-processing: since we will go through the supervised learning algorithms, we also need to program the ground truth. The ground truth of this research is programmed by comparing the closing price of the current trading date with the closing price of the previous trading date the users want to compare with. Label the price increase as 1, else the ground truth will be labeled as 0. Because this research work is not only focused on predicting the price trend of a specific period of time but short-term in general, the ground truth processing is according to a range of trading days. While the algorithms will not change with the prediction term length, we can regard the term length as a parameter.

The algorithmic detail is elaborated, respectively, the first algorithm is the hybrid feature engineering part for preparing high-quality training and testing data. It corresponds to the Feature extension, RFE, and PCA blocks in Fig.  3 . The second algorithm is the LSTM procedure block, including time-series data pre-processing, NN constructing, training, and testing.

Algorithm 1: Short-term stock market price trend prediction—applying feature engineering using FE + RFE + PCA

The function FE is corresponding to the feature extension block. For the feature extension procedure, we apply three different processing methods to translate the findings from the financial domain to a technical module in our system design. While not all the indices are applicable for expanding, we only choose the proper method(s) for certain features to perform the feature extension (FE), according to Table  2 .

Normalize method preserves the relative frequencies of the terms, and transform the technical indices into the range of [0, 1]. Polarize is a well-known method often used by real-world investors, sometimes they prefer to consider if the technical index value is above or below zero, we program some of the features using polarize method and prepare for RFE. Max-min (or min-max) [ 35 ] scaling is a transformation method often used as an alternative to zero mean and unit variance scaling. Another well-known method used is fluctuation percentage, and we transform the technical indices fluctuation percentage into the range of [− 1, 1].

The function RFE () in the first algorithm refers to recursive feature elimination. Before we perform the training data scale reduction, we will have to make sure that the features we selected are effective. Ineffective features will not only drag down the classification precision but also add more computational complexity. For the feature selection part, we choose recursive feature elimination (RFE). As [ 45 ] explained, the process of recursive feature elimination can be split into the ranking algorithm, resampling, and external validation.

For the ranking algorithm, it fits the model to the features and ranks by the importance to the model. We set the parameter to retain i numbers of features, and at each iteration of feature selection retains Si top-ranked features, then refit the model and assess the performance again to begin another iteration. The ranking algorithm will eventually determine the top Si features.

The RFE algorithm is known to have suffered from the over-fitting problem. To eliminate the over-fitting issue, we will run the RFE algorithm multiple times on randomly selected stocks as the training set and ensure all the features we select are high-weighted. This procedure is called data resampling. Resampling can be built as an optimization step as an outer layer of the RFE algorithm.

The last part of our hybrid feature engineering algorithm is for optimization purposes. For the training data matrix scale reduction, we apply Randomized principal component analysis (PCA) [ 31 ], before we decide the features of the classification model.

Financial ratios of a listed company are used to present the growth ability, earning ability, solvency ability, etc. Each financial ratio consists of a set of technical indices, each time we add a technical index (or feature) will add another column of data into the data matrix and will result in low training efficiency and redundancy. If non-relevant or less relevant features are included in training data, it will also decrease the precision of classification.

The above equation represents the explanation power of principal components extracted by PCA method for original data. If an ACR is below 85%, the PCA method would be unsuitable due to a loss of original information. Because the covariance matrix is sensitive to the order of magnitudes of data, there should be a data standardize procedure before performing the PCA. The commonly used standardized methods are mean-standardization and normal-standardization and are noted as given below:

Mean-standardization: \(X_{ij}^{*} = X_{ij} /\overline{{X_{j} }}\) , which \(\overline{{X_{j} }}\) represents the mean value.

Normal-standardization: \(X_{ij}^{*} = (X_{ij} - \overline{{X_{j} }} )/s_{j}\) , which \(\overline{{X_{j} }}\) represents the mean value, and \(s_{j}\) is the standard deviation.

The array fe_array is defined according to Table  2 , row number maps to the features, columns 0, 1, 2, 3 note for the extension methods of normalize, polarize, max–min scale, and fluctuation percentage, respectively. Then we fill in the values for the array by the rule where 0 stands for no necessity to expand and 1 for features need to apply the corresponding extension methods. The final algorithm of data preprocessing using RFE and PCA can be illustrated as Algorithm 1.

Algorithm 2: Price trend prediction model using LSTM

After the principal component extraction, we will get the scale-reduced matrix, which means i most effective features are converted into j principal components for training the prediction model. We utilized an LSTM model and added a conversion procedure for our stock price dataset. The detailed algorithm design is illustrated in Alg 2. The function TimeSeriesConversion () converts the principal components matrix into time series by shifting the input data frame according to the number of time steps [ 3 ], i.e., term length in this research. The processed dataset consists of the input sequence and forecast sequence. In this research, the parameter of LAG is 1, because the model is detecting the pattern of features fluctuation on a daily basis. Meanwhile, the N_TIME_STEPS is varied from 1 trading day to 10 trading days. The functions DataPartition (), FitModel (), EvaluateModel () are regular steps without customization. The NN structure design, optimizer decision, and other parameters are illustrated in function ModelCompile () .

Some procedures impact the efficiency but do not affect the accuracy or precision and vice versa, while other procedures may affect both efficiency and prediction result. To fully evaluate our algorithm design, we structure the evaluation part by main procedures and evaluate how each procedure affects the algorithm performance. First, we evaluated our solution on a machine with 2.2 GHz i7 processor, with 16 GB of RAM. Furthermore, we also evaluated our solution on Amazon EC2 instance, 3.1 GHz Processor with 16 vCPUs, and 64 GB RAM.

In the implementation part, we expanded 20 features into 54 features, while we retain 30 features that are the most effective. In this section, we discuss the evaluation of feature selection. The dataset was divided into two different subsets, i.e., training and testing datasets. Test procedure included two parts, one testing dataset is for feature selection, and another one is for model testing. We note the feature selection dataset and model testing dataset as DS_test_f and DS_test_m, respectively.

We randomly selected two-thirds of the stock data by stock ID for RFE training and note the dataset as DS_train_f; all the data consist of full technical indices and expanded features throughout 2018. The estimator of the RFE algorithm is SVR with linear kernels. We rank the 54 features by voting and get 30 effective features then process them using the PCA algorithm to perform dimension reduction and reduce the features into 20 principal components. The rest of the stock data forms the testing dataset DS_test_f to validate the effectiveness of principal components we extracted from selected features. We reformed all the data from 2018 as the training dataset of the data model and noted as DS_train_m. The model testing dataset DS_test_m consists of the first 3 months of data in 2019, which has no overlap with the dataset we utilized in the previous steps. This approach is to prevent the hidden problem caused by overfitting.

Term length

To build an efficient prediction model, instead of the approach of modeling the data to time series, we determined to use 1 day ahead indices data to predict the price trend of the next day. We tested the RFE algorithm on a range of short-term from 1 day to 2 weeks (ten trading days), to evaluate how the commonly used technical indices correlated to price trends. For evaluating the prediction term length, we fully expanded the features as Table  2 , and feed them to RFE. During the test, we found that different length of the term has a different level of sensitive-ness to the same indices set.

We get the close price of the first trading date and compare it with the close price of the n _ th trading date. Since we are predicting the price trend, we do not consider the term lengths if the cross-validation score is below 0.5. And after the test, as we can see from Fig.  4 , there are three-term lengths that are most sensitive to the indices we selected from the related works. They are n  = {2, 5, 10}, which indicates that price trend prediction of every other day, 1 week, and 2 weeks using the indices set are likely to be more reliable.

How do term lengths affect the cross-validation score of RFE

While these curves have different patterns, for the length of 2 weeks, the cross-validation score increases with the number of features selected. If the prediction term length is 1 week, the cross-validation score will decrease if selected over 8 features. For every other day price trend prediction, the best cross-validation score is achieved by selecting 48 features. Biweekly prediction requires 29 features to achieve the best score. In Table  3 , we listed the top 15 effective features for these three-period lengths. If we predict the price trend of every other day, the cross-validation score merely fluctuates with the number of features selected. So, in the next step, we will evaluate the RFE result for these three-term lengths, as shown in Fig.  4 .

We compare the output feature set of RFE with the all-original feature set as a baseline, the all-original feature set consists of n features and we choose n most effective features from RFE output features to evaluate the result using linear SVR. We used two different approaches to evaluate feature effectiveness. The first method is to combine all the data into one large matrix and evaluate them by running the RFE algorithm once. Another method is to run RFE for each individual stock and calculate the most effective features by voting.

Feature extension and RFE

From the result of the previous subsection, we can see that when predicting the price trend for every other day or biweekly, the best result is achieved by selecting a large number of features. Within the selected features, some features processed from extension methods have better ranks than original features, which proves that the feature extension method is useful for optimizing the model. The feature extension affects both precision and efficiency, while in this part, we only discuss the precision aspect and leave efficiency part in the next step since PCA is the most effective method for training efficiency optimization in our design. We involved an evaluation of how feature extension affects RFE and use the test result to measure the improvement of involving feature extension.

We further test the effectiveness of feature extension, i.e., if polarize, max–min scale, and calculate fluctuation percentage works better than original technical indices. The best case to leverage this test is the weekly prediction since it has the least effective feature selected. From the result we got from the last section, we know the best cross-validation score appears when selecting 8 features. The test consists of two steps, and the first step is to test the feature set formed by original features only, in this case, only SLOWK, SLOWD, and RSI_5 are included. The next step is to test the feature set of all 8 features we selected in the previous subsection. We leveraged the test by defining the simplest DNN model with three layers.

The normalized confusion matrix of testing the two feature sets are illustrated in Fig.  5 . The left one is the confusion matrix of the feature set with expanded features, and the right one besides is the test result of using original features only. Both precisions of true positive and true negative have been improved by 7% and 10%, respectively, which proves that our feature extension method design is reasonably effective.

Confusion matrix of validating feature extension effectiveness

Feature reduction using principal component analysis

PCA will affect the algorithm performance on both prediction accuracy and training efficiency, while this part should be evaluated with the NN model, so we also defined the simplest DNN model with three layers as we used in the previous step to perform the evaluation. This part introduces the evaluation method and result of the optimization part of the model from computational efficiency and accuracy impact perspectives.

In this section, we will choose bi-weekly prediction to perform a use case analysis, since it has a smoothly increasing cross-validation score curve, moreover, unlike every other day prediction, it has excluded more than 20 ineffective features already. In the first step, we select all 29 effective features and train the NN model without performing PCA. It creates a baseline of the accuracy and training time for comparison. To evaluate the accuracy and efficiency, we keep the number of the principal component as 5, 10, 15, 20, 25. Table  4 recorded how the number of features affects the model training efficiency, then uses the stack bar chart in Fig.  6 to illustrate how PCA affects training efficiency. Table  6 shows accuracy and efficiency analysis on different procedures for the pre-processing of features. The times taken shown in Tables  4 , 6 are based on experiments conducted in a standard user machine to show the viability of our solution with limited or average resource availability.

Relationship between feature number and training time

We also listed the confusion matrix of each test in Fig.  7 . The stack bar chart shows that the overall time spends on training the model is decreasing by the number of selected features, while the PCA method is significantly effective in optimizing training dataset preparation. For the time spent on the training stage, PCA is not as effective as the data preparation stage. While there is the possibility that the optimization effect of PCA is not drastic enough because of the simple structure of the NN model.

How does the number of principal components affect evaluation results

Table  5 indicates that the overall prediction accuracy is not drastically affected by reducing the dimension. However, the accuracy could not fully support if the PCA has no side effect to model prediction, so we looked into the confusion matrices of test results.

From Fig.  7 we can conclude that PCA does not have a severe negative impact on prediction precision. The true positive rate and false positive rate are barely be affected, while the false negative and true negative rates are influenced by 2% to 4%. Besides evaluating how the number of selected features affects the training efficiency and model performance, we also leveraged a test upon how data pre-processing procedures affect the training procedure and predicting result. Normalizing and max–min scaling is the most commonly seen data pre-procedure performed before PCA, since the measure units of features are varied, and it is said that it could increase the training efficiency afterward.

We leveraged another test on adding pre-procedures before extracting 20 principal components from the original dataset and make the comparison in the aspects of time elapse of training stage and prediction precision. However, the test results lead to different conclusions. In Table  6 we can conclude that feature pre-processing does not have a significant impact on training efficiency, but it does influence the model prediction accuracy. Moreover, the first confusion matrix in Fig.  8 indicates that without any feature pre-processing procedure, the false-negative rate and true negative rate are severely affected, while the true positive rate and false positive rate are not affected. If it performs the normalization before PCA, both true positive rate and true negative rate are decreasing by approximately 10%. This test also proved that the best feature pre-processing method for our feature set is exploiting the max–min scale.

Confusion matrices of different feature pre-processing methods

In this section, we discuss and compare the results of our proposed model, other approaches, and the most related works.

Comparison with related works

From the previous works, we found the most commonly exploited models for short-term stock market price trend prediction are support vector machine (SVM), multilayer perceptron artificial neural network (MLP), Naive Bayes classifier (NB), random forest classifier (RAF) and logistic regression classifier (LR). The test case of comparison is also bi-weekly price trend prediction, to evaluate the best result of all models, we keep all 29 features selected by the RFE algorithm. For MLP evaluation, to test if the number of hidden layers would affect the metric scores, we noted layer number as n and tested n  = {1, 3, 5}, 150 training epochs for all the tests, found slight differences in the model performance, which indicates that the variable of MLP layer number hardly affects the metric scores.

From the confusion matrices in Fig.  9 , we can see all the machine learning models perform well when training with the full feature set we selected by RFE. From the perspective of training time, training the NB model got the best efficiency. LR algorithm cost less training time than other algorithms while it can achieve a similar prediction result with other costly models such as SVM and MLP. RAF algorithm achieved a relatively high true-positive rate while the poor performance in predicting negative labels. For our proposed LSTM model, it achieves a binary accuracy of 93.25%, which is a significantly high precision of predicting the bi-weekly price trend. We also pre-processed data through PCA and got five principal components, then trained for 150 epochs. The learning curve of our proposed solution, based on feature engineering and the LSTM model, is illustrated in Fig.  10 . The confusion matrix is the figure on the right in Fig.  11 , and detailed metrics scores can be found in Table  9 .

Model prediction comparison—confusion matrices

Learning curve of proposed solution

Proposed model prediction precision comparison—confusion matrices

The detailed evaluate results are recorded in Table  7 . We will also initiate a discussion upon the evaluation result in the next section.

Because the resulting structure of our proposed solution is different from most of the related works, it would be difficult to make naïve comparison with previous works. For example, it is hard to find the exact accuracy number of price trend prediction in most of the related works since the authors prefer to show the gain rate of simulated investment. Gain rate is a processed number based on simulated investment tests, sometimes one correct investment decision with a large trading volume can achieve a high gain rate regardless of the price trend prediction accuracy. Besides, it is also a unique and heuristic innovation in our proposed solution, we transform the problem of predicting an exact price straight forward to two sequential problems, i.e., predicting the price trend first, focus on building an accurate binary classification model, construct a solid foundation for predicting the exact price change in future works. Besides the different result structure, the datasets that previous works researched on are also different from our work. Some of the previous works involve news data to perform sentiment analysis and exploit the SE part as another system component to support their prediction model.

The latest related work that can compare is Zubair et al. [ 47 ], the authors take multiple r-square for model accuracy measurement. Multiple r-square is also called the coefficient of determination, and it shows the strength of predictor variables explaining the variation in stock return [ 28 ]. They used three datasets (KSE 100 Index, Lucky Cement Stock, Engro Fertilizer Limited) to evaluate the proposed multiple regression model and achieved 95%, 89%, and 97%, respectively. Except for the KSE 100 Index, the dataset choice in this related work is individual stocks; thus, we choose the evaluation result of the first dataset of their proposed model.

We listed the leading stock price trend prediction model performance in Table  8 , from the comparable metrics, the metric scores of our proposed solution are generally better than other related works. Instead of concluding arbitrarily that our proposed model outperformed other models in related works, we first look into the dataset column of Table  8 . By looking into the dataset used by each work [ 18 ], only trained and tested their proposed solution on three individual stocks, which is difficult to prove the generalization of their proposed model. Ayo [ 2 ] leveraged analysis on the stock data from the New York Stock Exchange (NYSE), while the weakness is they only performed analysis on closing price, which is a feature embedded with high noise. Zubair et al. [ 47 ] trained their proposed model on both individual stocks and index price, but as we have mentioned in the previous section, index price only consists of the limited number of features and stock IDs, which will further affect the model training quality. For our proposed solution, we collected sufficient data from the Chinese stock market, and applied FE + RFE algorithm on the original indices to get more effective features, the comprehensive evaluation result of 3558 stock IDs can reasonably explain the generalization and effectiveness of our proposed solution in Chinese stock market. However, the authors of Khaidem and Dey [ 18 ] and Ayo [ 2 ] chose to analyze the stock market in the United States, Zubair et al. [ 47 ] performed analysis on Pakistani stock market price, and we obtained the dataset from Chinese stock market, the policies of different countries might impact the model performance, which needs further research to validate.

Proposed model evaluation—PCA effectiveness

Besides comparing the performance across popular machine learning models, we also evaluated how the PCA algorithm optimizes the training procedure of the proposed LSTM model. We recorded the confusion matrices comparison between training the model by 29 features and by five principal components in Fig.  11 . The model training using the full 29 features takes 28.5 s per epoch on average. While it only takes 18 s on average per epoch training on the feature set of five principal components. PCA has significantly improved the training efficiency of the LSTM model by 36.8%. The detailed metrics data are listed in Table  9 . We will leverage a discussion in the next section about complexity analysis.

Complexity analysis of proposed solution

This section analyzes the complexity of our proposed solution. The Long Short-term Memory is different from other NNs, and it is a variant of standard RNN, which also has time steps with memory and gate architecture. In the previous work [ 46 ], the author performed an analysis of the RNN architecture complexity. They introduced a method to regard RNN as a directed acyclic graph and proposed a concept of recurrent depth, which helps perform the analysis on the intricacy of RNN.

The recurrent depth is a positive rational number, and we denote it as \(d_{rc}\) . As the growth of \(n\) \(d_{rc}\) measures, the nonlinear transformation average maximum number of each time step. We then unfold the directed acyclic graph of RNN and denote the processed graph as \(g_{c}\) , meanwhile, denote \(C(g_{c} )\) as the set of directed cycles in this graph. For the vertex \(v\) , we note \(\sigma_{s} (v)\) as the sum of edge weights and \(l(v)\) as the length. The equation below is proved under a mild assumption, which could be found in [ 46 ].

They also found that another crucial factor that impacts the performance of LSTM, which is the recurrent skip coefficients. We note \(s_{rc}\) as the reciprocal of the recurrent skip coefficient. Please be aware that \(s_{rc}\) is also a positive rational number.

According to the above definition, our proposed model is a 2-layers stacked LSTM, which \(d_{rc} = 2\) and \(s_{rc} = 1\) . From the experiments performed in previous work, the authors also found that when facing the problems of long-term dependency, LSTMs may benefit from decreasing the reciprocal of recurrent skip coefficients and from increasing recurrent depth. The empirical findings above mentioned are useful to enhance the performance of our proposed model further.

This work consists of three parts: data extraction and pre-processing of the Chinese stock market dataset, carrying out feature engineering, and stock price trend prediction model based on the long short-term memory (LSTM). We collected, cleaned-up, and structured 2 years of Chinese stock market data. We reviewed different techniques often used by real-world investors, developed a new algorithm component, and named it as feature extension, which is proved to be effective. We applied the feature expansion (FE) approaches with recursive feature elimination (RFE), followed by principal component analysis (PCA), to build a feature engineering procedure that is both effective and efficient. The system is customized by assembling the feature engineering procedure with an LSTM prediction model, achieved high prediction accuracy that outperforms the leading models in most related works. We also carried out a comprehensive evaluation of this work. By comparing the most frequently used machine learning models with our proposed LSTM model under the feature engineering part of our proposed system, we conclude many heuristic findings that could be future research questions in both technical and financial research domains.

Our proposed solution is a unique customization as compared to the previous works because rather than just proposing yet another state-of-the-art LSTM model, we proposed a fine-tuned and customized deep learning prediction system along with utilization of comprehensive feature engineering and combined it with LSTM to perform prediction. By researching into the observations from previous works, we fill in the gaps between investors and researchers by proposing a feature extension algorithm before recursive feature elimination and get a noticeable improvement in the model performance.

Though we have achieved a decent outcome from our proposed solution, this research has more potential towards research in future. During the evaluation procedure, we also found that the RFE algorithm is not sensitive to the term lengths other than 2-day, weekly, biweekly. Getting more in-depth research into what technical indices would influence the irregular term lengths would be a possible future research direction. Moreover, by combining latest sentiment analysis techniques with feature engineering and deep learning model, there is also a high potential to develop a more comprehensive prediction system which is trained by diverse types of information such as tweets, news, and other text-based data.

Abbreviations

Long short term memory

Principal component analysis

Recurrent neural networks

Artificial neural network

Deep neural network

Dynamic Time Warping

Recursive feature elimination

Support vector machine

Convolutional neural network

Stochastic gradient descent

Rectified linear unit

Multi layer perceptron

Atsalakis GS, Valavanis KP. Forecasting stock market short-term trends using a neuro-fuzzy based methodology. Expert Syst Appl. 2009;36(7):10696–707.

Article   Google Scholar  

Ayo CK. Stock price prediction using the ARIMA model. In: 2014 UKSim-AMSS 16th international conference on computer modelling and simulation. 2014. https://doi.org/10.1109/UKSim.2014.67 .

Brownlee J. Deep learning for time series forecasting: predict the future with MLPs, CNNs and LSTMs in Python. Machine Learning Mastery. 2018. https://machinelearningmastery.com/time-series-prediction-lstm-recurrent-neural-networks-python-keras/

Eapen J, Bein D, Verma A. Novel deep learning model with CNN and bi-directional LSTM for improved stock market index prediction. In: 2019 IEEE 9th annual computing and communication workshop and conference (CCWC). 2019. pp. 264–70. https://doi.org/10.1109/CCWC.2019.8666592 .

Fischer T, Krauss C. Deep learning with long short-term memory networks for financial market predictions. Eur J Oper Res. 2018;270(2):654–69. https://doi.org/10.1016/j.ejor.2017.11.054 .

Article   MathSciNet   MATH   Google Scholar  

Guyon I, Weston J, Barnhill S, Vapnik V. Gene selection for cancer classification using support vector machines. Mach Learn 2002;46:389–422.

Hafezi R, Shahrabi J, Hadavandi E. A bat-neural network multi-agent system (BNNMAS) for stock price prediction: case study of DAX stock price. Appl Soft Comput J. 2015;29:196–210. https://doi.org/10.1016/j.asoc.2014.12.028 .

Halko N, Martinsson PG, Tropp JA. Finding structure with randomness: probabilistic algorithms for constructing approximate matrix decompositions. SIAM Rev. 2001;53(2):217–88.

Article   MathSciNet   Google Scholar  

Hassan MR, Nath B. Stock market forecasting using Hidden Markov Model: a new approach. In: Proceedings—5th international conference on intelligent systems design and applications 2005, ISDA’05. 2005. pp. 192–6. https://doi.org/10.1109/ISDA.2005.85 .

Hochreiter S, Schmidhuber J. Long short-term memory. J Neural Comput. 1997;9(8):1735–80. https://doi.org/10.1162/neco.1997.9.8.1735 .

Hsu CM. A hybrid procedure with feature selection for resolving stock/futures price forecasting problems. Neural Comput Appl. 2013;22(3–4):651–71. https://doi.org/10.1007/s00521-011-0721-4 .

Huang CF, Chang BR, Cheng DW, Chang CH. Feature selection and parameter optimization of a fuzzy-based stock selection model using genetic algorithms. Int J Fuzzy Syst. 2012;14(1):65–75. https://doi.org/10.1016/J.POLYMER.2016.08.021 .

Huang CL, Tsai CY. A hybrid SOFM-SVR with a filter-based feature selection for stock market forecasting. Expert Syst Appl. 2009;36(2 PART 1):1529–39. https://doi.org/10.1016/j.eswa.2007.11.062 .

Idrees SM, Alam MA, Agarwal P. A prediction approach for stock market volatility based on time series data. IEEE Access. 2019;7:17287–98. https://doi.org/10.1109/ACCESS.2019.2895252 .

Ince H, Trafalis TB. Short term forecasting with support vector machines and application to stock price prediction. Int J Gen Syst. 2008;37:677–87. https://doi.org/10.1080/03081070601068595 .

Jeon S, Hong B, Chang V. Pattern graph tracking-based stock price prediction using big data. Future Gener Comput Syst. 2018;80:171–87. https://doi.org/10.1016/j.future.2017.02.010 .

Kara Y, Acar Boyacioglu M, Baykan ÖK. Predicting direction of stock price index movement using artificial neural networks and support vector machines: the sample of the Istanbul Stock Exchange. Expert Syst Appl. 2011;38(5):5311–9. https://doi.org/10.1016/j.eswa.2010.10.027 .

Khaidem L, Dey SR. Predicting the direction of stock market prices using random forest. 2016. pp. 1–20.

Kim K, Han I. Genetic algorithms approach to feature discretization in artificial neural networks for the prediction of stock price index. Expert Syst Appl. 2000;19:125–32. https://doi.org/10.1016/S0957-4174(00)00027-0 .

Lee MC. Using support vector machine with a hybrid feature selection method to the stock trend prediction. Expert Syst Appl. 2009;36(8):10896–904. https://doi.org/10.1016/j.eswa.2009.02.038 .

Lei L. Wavelet neural network prediction method of stock price trend based on rough set attribute reduction. Appl Soft Comput J. 2018;62:923–32. https://doi.org/10.1016/j.asoc.2017.09.029 .

Lin X, Yang Z, Song Y. Expert systems with applications short-term stock price prediction based on echo state networks. Expert Syst Appl. 2009;36(3):7313–7. https://doi.org/10.1016/j.eswa.2008.09.049 .

Liu G, Wang X. A new metric for individual stock trend prediction. Eng Appl Artif Intell. 2019;82(March):1–12. https://doi.org/10.1016/j.engappai.2019.03.019 .

Liu S, Zhang C, Ma J. CNN-LSTM neural network model for quantitative strategy analysis in stock markets. 2017;1:198–206. https://doi.org/10.1007/978-3-319-70096-0 .

Long W, Lu Z, Cui L. Deep learning-based feature engineering for stock price movement prediction. Knowl Based Syst. 2018;164:163–73. https://doi.org/10.1016/j.knosys.2018.10.034 .

Malkiel BG, Fama EF. Efficient capital markets: a review of theory and empirical work. J Finance. 1970;25(2):383–417.

McNally S, Roche J, Caton S. Predicting the price of bitcoin using machine learning. In: Proceedings—26th Euromicro international conference on parallel, distributed, and network-based processing, PDP 2018. pp. 339–43. https://doi.org/10.1109/PDP2018.2018.00060 .

Nagar A, Hahsler M. News sentiment analysis using R to predict stock market trends. 2012. http://past.rinfinance.com/agenda/2012/talk/Nagar+Hahsler.pdf . Accessed 20 July 2019.

Nekoeiqachkanloo H, Ghojogh B, Pasand AS, Crowley M. Artificial counselor system for stock investment. 2019. ArXiv Preprint arXiv:1903.00955 .

Ni LP, Ni ZW, Gao YZ. Stock trend prediction based on fractal feature selection and support vector machine. Expert Syst Appl. 2011;38(5):5569–76. https://doi.org/10.1016/j.eswa.2010.10.079 .

Pang X, Zhou Y, Wang P, Lin W, Chang V. An innovative neural network approach for stock market prediction. J Supercomput. 2018. https://doi.org/10.1007/s11227-017-2228-y .

Pimenta A, Nametala CAL, Guimarães FG, Carrano EG. An automated investing method for stock market based on multiobjective genetic programming. Comput Econ. 2018;52(1):125–44. https://doi.org/10.1007/s10614-017-9665-9 .

Piramuthu S. Evaluating feature selection methods for learning in data mining applications. Eur J Oper Res. 2004;156(2):483–94. https://doi.org/10.1016/S0377-2217(02)00911-6 .

Qiu M, Song Y. Predicting the direction of stock market index movement using an optimized artificial neural network model. PLoS ONE. 2016;11(5):e0155133.

Scikit-learn. Scikit-learn Min-Max Scaler. 2019. https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.MinMaxScaler.html . Retrieved 26 July 2020.

Shen J. Thesis, “Short-term stock market price trend prediction using a customized deep learning system”, supervised by M. Omair Shafiq, Carleton University. 2019.

Shen J, Shafiq MO. Deep learning convolutional neural networks with dropout—a parallel approach. ICMLA. 2018;2018:572–7.

Google Scholar  

Shen J, Shafiq MO. Learning mobile application usage—a deep learning approach. ICMLA. 2019;2019:287–92.

Shih D. A study of early warning system in volume burst risk assessment of stock with Big Data platform. In: 2019 IEEE 4th international conference on cloud computing and big data analysis (ICCCBDA). 2019. pp. 244–8.

Sirignano J, Cont R. Universal features of price formation in financial markets: perspectives from deep learning. Ssrn. 2018. https://doi.org/10.2139/ssrn.3141294 .

Article   MATH   Google Scholar  

Thakur M, Kumar D. A hybrid financial trading support system using multi-category classifiers and random forest. Appl Soft Comput J. 2018;67:337–49. https://doi.org/10.1016/j.asoc.2018.03.006 .

Tsai CF, Hsiao YC. Combining multiple feature selection methods for stock prediction: union, intersection, and multi-intersection approaches. Decis Support Syst. 2010;50(1):258–69. https://doi.org/10.1016/j.dss.2010.08.028 .

Tushare API. 2018. https://github.com/waditu/tushare . Accessed 1 July 2019.

Wang X, Lin W. Stock market prediction using neural networks: does trading volume help in short-term prediction?. n.d.

Weng B, Lu L, Wang X, Megahed FM, Martinez W. Predicting short-term stock prices using ensemble methods and online data sources. Expert Syst Appl. 2018;112:258–73. https://doi.org/10.1016/j.eswa.2018.06.016 .

Zhang S. Architectural complexity measures of recurrent neural networks, (NIPS). 2016. pp. 1–9.

Zubair M, Fazal A, Fazal R, Kundi M. Development of stock market trend prediction system using multiple regression. Computational and mathematical organization theory. Berlin: Springer US; 2019. https://doi.org/10.1007/s10588-019-09292-7 .

Book   Google Scholar  

Download references

Acknowledgements

This research is supported by Carleton University, in Ottawa, ON, Canada. This research paper has been built based on the thesis [ 36 ] of Jingyi Shen, supervised by M. Omair Shafiq at Carleton University, Canada, available at https://curve.carleton.ca/52e9187a-7f71-48ce-bdfe-e3f6a420e31a .

NSERC and Carleton University.

Author information

Authors and affiliations.

School of Information Technology, Carleton University, Ottawa, ON, Canada

Jingyi Shen & M. Omair Shafiq

You can also search for this author in PubMed   Google Scholar

Contributions

Yes. All authors read and approved the final manuscript.

Corresponding author

Correspondence to M. Omair Shafiq .

Ethics declarations

Competing interests.

The authors declare that they have no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Shen, J., Shafiq, M.O. Short-term stock market price trend prediction using a comprehensive deep learning system. J Big Data 7 , 66 (2020). https://doi.org/10.1186/s40537-020-00333-6

Download citation

Received : 24 January 2020

Accepted : 30 July 2020

Published : 28 August 2020

DOI : https://doi.org/10.1186/s40537-020-00333-6

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Deep learning
• Stock market trend
• Feature engineering

Macroeconomics and Finance: The Role of the Stock Market

The treatment of the stock market in finance and macroeconomics exemplifies many of the important differences in perspective between the two fields. In finance, the stock market is the single most important market with respect to corporate investment decisions. In contrast, macroeconomic modelling and policy discussion assign a relatively minor role to the stockmarket in investment decisions. This paper explores four possible explanations for this neglect and concludes that macro analysis should give more attention to the stock market. Despite the frequent jibe that "the stockmarket has forecast ten of the last six recessions," the stock market is in fact a good predictor of the business cycle and the components of GNP. We examine the relative importance of the required return on equity compared with the interest rate in the determination of the cost of capital, and hence,investment. In this connection, we review the empirical success of the Q theory of investment which relates investment to stock market evaluations of firms. One of the explanations for the neglect of the stock market in macroeconomics may be the view that because the stock market fluctuates excessively, rational managers will pay little attention to the market informulating investment plans. This view is shown to be unfounded by demonstrating that rational managers will react to stock price changes even if the stock market fluctuates excessively. Finally, we review the extremely important issue of whether the market does fluctuate excessively, and conclude that while not ruled out on a priori theoretical grounds, the empirical evidence for such excess fluctuations has not been decisive.

• Acknowledgements and Disclosures

MARC RIS BibTeΧ

Download Citation Data

Published Versions

More from nber.

In addition to working papers , the NBER disseminates affiliates’ latest findings through a range of free periodicals — the NBER Reporter , the NBER Digest , the Bulletin on Retirement and Disability , the Bulletin on Health , and the Bulletin on Entrepreneurship  — as well as online conference reports , video lectures , and interviews .

Reading Lists +

The review +, 46 possible stock market strategies from academics get a retest.

9 March 2022

Research by

• Athanasse Zafirov
• Corporate Investment
• Stock Market
• Corporate Finance
• Stock Returns

We won’t call it debunking, but not all investing tips hold up

For nearly 3,000 years, bloodletting was an accepted medical practice for all types of maladies. It was only in the early 1800s when some doctors carefully reviewed data on the practice that they realized bloodletting didn’t improve patients’ health, and may sometimes be harmful.

Such review of accepted theories is currently a growing field among social and natural scientists. Peer-reviewed research is increasingly being thrown back into the review process to see if stands up.

Opt In to the Review Monthly Email Update.

A working paper by the University of Lausanne’s Amit Goyal, UCLA Anderson’s Ivo Welch and Athanasse Zafirov, a Ph.D. student, seeks to prevent the financial equivalent of bloodletting. Their meta-research — the term given for research on research — on papers published in top academic journals finds that many investing factors don’t hold up. To be precise, the 46 variables aren’t full-blown market strategies, but rather observed correlations that could form the basis for a strategy.

Past Performance May Not Be Indicative of Future Results

Building on Goyal and Welch’s 2008 paper that studied the predictive success of 17 variables, the researchers survey 26 papers identifying 29 variables considered useful in predicting the equity premium — the total rate of return on the stock market minus the prevailing short-term interest rate. The 17 variables from the 2008 paper are also reexamined. The researchers’ findings suggest that most of the variables have lost their predictive ability when tested on datasets extended to the end of 2020. A few variables do show flickers of promise but not overwhelming success across the researchers’ evaluation metrics.

The researchers’ first goal was to replicate the original findings of the papers’ authors. This involved recreating the variables and recalculating the reported statistics on the variables’ ability to predict the equity premium. Goyal, Welch and Zafirov were able to confirm the papers’ original findings, using the original dataset, on all but two of the papers. (The two remaining papers had data issues.)

The datasets to create the variables were then extended through December 2020, and the predictions for each of the 29 variables from the papers and the original 17 variables from the 2008 paper were retested.

The datasets in the papers ended between 2000 and 2017 and began as early as 1926. When building a predictive model, a researcher will typically split a dataset into at least two samples —one sample to train the model and another sample, typically the data from the latest years, to test the model. By extending the original datasets with data to the end of 2020 and starting the test sample 20 years after the start of the training sample, the components of these samples were slightly different than the samples used in the papers. It’s worth noting that the new data only made up a small percentage of the overall datasets.

“Because our paper reuses the data that the authors themselves had originally used to discover and validate their variables and theories, all that the predictors had to do in the few added years was not to ‘screw up’ badly.”

Nonetheless, of the 46 variables, only five managed to predict at a statistically significant level on the samples in the extended dataset.

But statistics are one thing, and investment performance is another. As a second test, Goyal, Welch and Zafirov devised simple investment strategies using the variables’ predictions to time investments by determining whether to go long or short the market and weighting the investments. The results of the investment strategies were compared with a buy-and-hold strategy. None of the five variables was able to significantly outperform the buy-and-hold approach in any of the investment strategies. Across all of the variable predictors, half lost money in the simplest investment strategy that used the variable to determine whether to go long or short.

Why Does the Performance Degrade?

The researchers suggest that the deterioration in predictive performance is at least partly explained by the fact that the market has shown greater variety in regimes over the last 20 years with many steep downturns. Campbell R. Harvey of Duke University and Yan Liu of Purdue University have performed similar meta-research and suggest that over-adapting the model to a particular data set may also be a factor due to authors running numerous backtests (simulations over historical data); they further suggest increasing necessary performance thresholds (raising the bar) as the number of backtests increase. Finally, a more generous explanation may be that as the predictive variables become well known by market practitioners, they lose their edge, just like a stock tip — when those tipped off start buying, the stock price rises and the tip loses its value.

Looking at the table below, the variables that were found to remain statistically significant on the extended dataset were those with the fewest citations and likely less well known among market participants.

The Five Best Variables on a Statistical Basis

Fourth-Quarter Growth Rate in Personal Consumption Expenditures ( gpce) : This macroeconomic variable from researchers Møller and Rangvid posits that high personal consumption growth rates at the end of the year predicts poor stock-market gains in the following year. The researchers found it to be the best, and most consistent, variable in the investment strategies. It outperformed a buy-and-hold approach with three of the four strategies tested. However, the outperformance was only marginal.

Aggregate Accruals (accru) : This is a sentiment-based variable introduced by Hirshleifer, Hou and Teoh and uses aggressive corporate accounting to predict future stock returns — more aggressive accruals lead to lower future returns. The variable also marginally beat buy-and-hold returns in three out of four approaches. Most of its performance came from its prediction of the post-tech market crash in 2000-2002.

Credit Standards (crdstd) : This is another macroeconomic variable and was introduced by Chava, Gallmeyer and Park. It finds that optimistic (loose) credit standards predict poor market returns and comes from survey data by the Fed. This variable did well in the researchers’ investment strategies and had good performance on test sample data, but statistical measures of the variable on the training sample data were not as convincing and much of its performance comes from the first four years in that sample.

The Investment Capital Ratio (i/k) : This a financial ratio introduced by Cochrane all the way back in 1991 and was also included in the 2008 paper from Goyal and Welch. It posits that high capital investment in the current quarter predicts poor stock-market returns in the next quarter. While it was a poor predictor from 1975 to 1998, it has since improved performance yet was not able to outperform a buy-and-hold strategy in three of four of the researchers’ timing strategies.

Treasury-bill Rates (tbl ): This is another variable examined in the 2008 paper. It does well statistically but had poor performance in the investment strategies.

Oft-Cited Papers With Poor-Performing Variables

Variance Risk Premium (vrp) : This variable was introduced by Bollerslev, Tauchen and Zhou and has the most citations. The variable had poor statistical performance, as well as poor performance in all four of the investment strategies.

Share of Housing Consumption (house) : This macroeconomic variable introduced by Piazzesi, Schneider and Tuzel has the second-highest number of citations. It uses housing share of consumer spending to forecast the excess return of stocks. (The higher the spending on housing, the higher the excess returns in the stock market.) The variable had poor statistical performance on the extended dataset and poor performance in the investment strategies.

The Price of West-Texas Intermediate Crude Oil (wtexas) : This was the only commodity-based variable and was introduced by Driesprong, Jacobsen and Maat. The paper posits that changes in the price of oil predict stock returns — higher oil prices lead to lower stock returns — with lags. The variable had poor statistical performance for the extended dataset and inconsistent performance in the investment strategies.

The First Principal Component of 14 Technical Indicators ( tchi ): This variable was introduced by Neely, Rapach, Tu and Zhou and is a linear combination of technical indicators including moving price averages, momentum and volume. It only had marginal statistical performance and inconsistent performance in the trading strategies.

Featured Faculty

Distinguished Professor of Finance; J. Fred Weston Chair in Finance

About the Research

Goyal, A., Welch, I., & Zafirov, A. (2021). A Comprehensive Look at the Empirical Performance of Equity Premium Prediction II . http://dx.doi.org/10.2139/ssrn.3929119

Suggested Articles

New Study Disavows Marshmallow Test’s Predictive Powers

Replicating a Successful Nudge in Health Care: Advice for Skeptics

Calculating a Value for the Government Support Banks Enjoy

Related articles.

A Different Way to Evaluate Private Equity Performance

An approach tailored to investor risk appetite and more comparable to stocks

A Solution to the Debate over Momentum’s Cause?

Investors may underreact when information arrives in small, continuous bits

When Younger Investors Overreact to News, Others Feel It

Inexperienced investors, lacking historical context, impact markets

The Strange Case of the Missing Stock Market Return

Investors in leveraged companies take on extra risk, but research indicates they see no offsetting return

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Wiley - PMC COVID-19 Collection

The effect of COVID‐19 on the global stock market

Pattanaporn chatjuthamard.

1 Sasin Graduate Institute of Business Administration of Chulalongkorn University, Bangkok Thailand

Pavitra Jindahra

Pattarake sarajoti, sirimon treepongkaruna.

2 UWA Business School, University of Western Australia, Perth WA, Australia

This paper investigates the effect of COVID‐19 on the global stock market. Specifically, we test whether the growth in the number of confirmed cases/deaths affects market quality, measured by return, realised volatility, jumps and co‐jumps for 43 stock indices around the world. We find that an increase in the growth rate of the number of confirmed cases increases volatility and jumps while reducing return. Further, we explore whether economic, financial and political risks play any significant role in the relation between the number of confirmed cases/deaths and market quality. Overall, we find the risk from COVID‐19 overshadows these risks.

1. Introduction

In just a matter of weeks, the contagious virus COVID‐19 spread around the world, leading to a global pandemic and destructive economic impacts on an unparalleled scale (see Baldwin and Di Mauro, 2020 ; Goodell, 2020 ). Despite extensive research related to COVID‐19, our understanding of COVID‐19 and its effects on market quality are still relatively limited. The outbreak of COVID‐19 caused more frequent stock market index jumps than any other period in history with the same number of trading days (Baker et al ., 2020 ). Yet, there is minimal scrutiny of the impact of the pandemic on jumps and co‐jumps of stock indices. This paper fills this gap in the existing literature by investigating how COVID‐19 affects returns, volatility, and jumps of the stock market indices. Further, it explores whether COVID‐19 causes global stock market indices to jump with the S&P500 index. Finally, it assesses whether country risk improves or impairs the above relationships.

This new infectious disease is distinct from and much more dangerous than previous outbreaks (Alfaro et al ., 2020 ; Baker et al ., 2020 ; Jackwerth, 2020 ). Not only does it lower market returns, it also increases the volatility of the stock market (Al‐Awadhi et al ., 2020 ; Ashraf, 2020 ; Erdem, 2020 ; Ramelli and Wagner, 2020 ). Nevertheless, the impact of COVID‐19 has caused investors to suffer significant losses in a short period of time due to a very high level of risks (Zhang et al ., 2020 ). Although the COVID‐19 shock has been global, not all countries have been impacted in the same way, and they have not reacted in the same way. Some researchers identify firm characteristics which soften the adverse effects of the health crisis (Albuquerque et al ., 2020 ; Fahlenbrach et al ., 2020 ; Ramelli and Wagner, 2020 ), while others suggest political and social progress are key determinants in explaining the heterogeneous impacts of COVID‐19 on stock returns across countries (Greer et al ., 2020 ).

Stock volatility is not directly observable, but rather inherently latent. In response, several studies, such as Andersen et al . ( 2010 ), Andersen et al . ( 2011 ) and Phiromswad et al . ( 2021 ), advocate the use of so‐called realised volatilities (constructed from the summation of the squared intraday interval return) as a practical method for improving the ex‐post volatility measures. Theoretically, these realised volatilities are free from measurement error (Andersen et al ., 2003 ). In addition, Andersen et al . ( 2003 ) indicate that simple models of realised volatility outperform the well‐known GARCH and related stochastic volatility models in out‐of‐sample forecasting. In our analysis, we separate the realised volatility into continuous and discontinuous jump components by using the nonparametric techniques developed by Barndorff‐Nielsen and Shephard ( 2004 ). These components correspond to the expected and unexpected new events. Prior research indicates that financial market jumps are responsible for the majority of market volatility, especially during crisis periods (Chan et al ., 2014 ).

We contribute to the literature in several respects. First, we add to the literature that utilises high‐frequency data to capture volatility dynamics (e.g., Andersen et al ., 2003 ; Tanthanongsakkun et al ., 2018 ; Ho et al ., 2021 ; Phiromswad et al ., 2021 ). For instance, Phiromswad et al . ( 2021 ) examine co‐jumps of 54 cryptocurrencies with the Thai stock market. Wang et al . ( 2020 ) investigate the usefulness of the implied volatility index (VIX) and the economic policy uncertainty (EPU) index in forecasting future volatility for 19 equity indices, finding that the VIX is a better predictor than the EPU index during the coronavirus pandemic. Chan et al . ( 2014 ) examine whether currency jumps are more severe in emerging markets, especially during crises, while Dungey et al . ( 2014 ) use high frequency data to detect stress dates in currency markets. Our analysis complements the financial market studies of Wang et al . ( 2020 ), Chan et al . ( 2014 ) and Dungey et al . ( 2014 ), who examine the impact of crises on volatility dynamics. More specifically, by decomposing volatility into continuous and discontinuous jump components, this paper provides a novel way to understand the impact of COVID‐19 on the volatility of stock indices. This method is also less vulnerable to market microstructure noise, which is a key concern in the asset pricing literature (Andersen et al ., 2007 ).

Our findings also relate to the impact of the pandemic on the co‐movements of global markets (e.g., Akhtaruzzaman et al ., 2020 ; He et al ., 2020 ; Okorie and Lin, 2020 ). Akhtaruzzaman et al . ( 2020 ) report a significant increase in stock market correlations between China and G7 countries during the pandemic period. Similarly, He et al . ( 2020 ) document that the impact of COVID‐19 on the European and US stock markets has a backflow effect on the Asian stock markets, particularly China. Distinct from these studies, we exploit the intraday 5‐min return to construct co‐jumps of the US stock index and other stock market indices around the world.

As such, our paper is part of the emerging literature which examines the impact of COVID‐19 on financial outcomes (e.g., Al‐Awadhi et al ., 2020 ; Ashraf, 2020 ; Erdem, 2020 ; Ramelli and Wagner, 2020 ). Alfaro et al . ( 2020 ) examine the relationship between unanticipated changes in COVID‐19 infections and aggregate market returns. Baker et al . ( 2020 ) and Zaremba et al . ( 2020 ) investigate the effect of government interventions in contributing to stock market volatility. Building on this literature, our study provides novel evidence of the heterogeneous impacts of COVID‐19 on stock returns across countries.

Based on 43 5‐min intraday stock indices over the period 30 October 2019–13 May 2020, our results suggest that the COVID‐19 pandemic has exerted a negative and significant impact on market quality across the globe. In particular, we show that the pandemic negatively affects stock market returns but positively affects stock market volatility, jumps and co‐jumps. Furthermore, there is weak evidence suggesting that country risk lowers the impact of COVID‐19 on market quality.

The remainder of this paper is organised as follows. Section  2 reviews related literature and develops hypotheses. Section  3 describes the data and methodology. Section 4 presents empirical results and Section  5 concludes.

2. Literature review and hypothesis development

The efficient market hypothesis (EMH) assumes that all investors are rational and stock prices adequately reflect all available information. However, many financial anomalies (such as excess volatility and systemic under‐ or over‐valuation of stock prices relative to their intrinsic values) cannot be explained by the EMH. Behavioural finance researchers believe investor sentiment may help to explain these market anomalies. According to Black ( 1986 ) and De Long et al . ( 1990 ), there are two types of investors: informed rational investors and noise traders. Rational informed investors, who are sentiment free, form rational expectations about the expected future cash flow of asset values. In contrast, uninformed noise traders experience waves of irrational sentiment and tend to form cognitive bias expectations, causing strong and persistent mispricing. Both types of investors compete in the market and set prices and expected returns; hence, the equilibrium price reflects the opinions of both rational investors and noise traders.

External and unexpected shocks, such as a financial crisis or disease outbreak, can affect economic trends and suddenly change investors' sentiments. When the market is trending downward, investors behave more pessimistically, leading to upward revisions in volatility and lower future excess returns (Lee et al ., 2002 ). Burns et al . ( 2012 ) suggest that perceived risk and negative emotions often escalate in the initial stage of a crisis as the public responds to news reports, social media and social interaction with friends and family. Along the same line, Roszkowski and Davey ( 2010 ) document the dramatic increase in the public's perception of the risk inherent in investing during the financial crisis of 2008.

The impact of investor sentiment on the stock market during a crisis is well documented. Several empirical studies rely on VIX as a proxy for the overall attitude or tone of investors towards future cash flows and investment risk of a particular security or financial market (see, e.g., Altig et al ., 2020 ; Cheng, 2020 ; Jackwerth, 2020 ). A rising VIX implies an increased need for protection against risk and is a sign of increasing market volatility; in particular, VIX is used as a measurement of investors' fear. Other researchers focus on implied volatility from stochastic volatility models (see, e.g., Alan et al ., 2020 ; Mirza et al ., 2020 ). Nevertheless, in practice, stock volatility is not directly observable. Andersen et al . ( 2001 ) and others suggest the use of so‐called realised volatilities, constructed from the summation of the squared intraday interval return, as a practical method for improving the ex‐post volatility measures. It is free from measurement error and outperforms the well‐known GARCH and related stochastic volatility models in out‐of‐sample forecasting (Andersen et al ., 2003 ).

COVID‐19 is much more than a health crisis; it is also very much an economic crisis that has affected the lives of many individuals, families and businesses across various industries globally. The global financial markets reacted very strongly and stock market returns dropped sharply as the COVID‐19 pandemic grew (Al‐Awadhi et al ., 2020 ; Ashraf, 2020 ; Erdem, 2020 ; Ramelli and Wagner, 2020 ). However, the impact of the increasing number of deaths on the stock market remains unclear. 1

As more and more cases were diagnosed, investors became wary about the unusual uncertainty surrounding the financial markets, leading to a highly volatile and unpredictable market situation. Baker et al . ( 2020 ) note that from 24 February to 24 March 2020, there were 18 market jumps, largely due to reactions to news about COVID‐19 in the United States. Alfaro et al . ( 2020 ) show that US stock returns respond to daily unanticipated changes in COVID‐19 infections, implying declining stock market volatility as the pandemic became less uncertain. Alan et al . ( 2020 ) and Zaremba et al . ( 2020 ) demonstrate the impact of governments' policy response to the pandemic on stock market volatility. Similarly, Zaremba et al . ( 2021a ) investigate the role of non‐pharmaceutical interventions in equity market liquidity. Finally, Zhang et al . ( 2020 ) find that the number of COVID‐19 confirmed cases causes an increase in country‐specific risks in stock markets as well as systemic risks.

Globalisation has linked global economies and increased the interdependence of global financial markets. Akhtaruzzaman et al . ( 2020 ) show that listed firms across China and G7 countries have experienced significant increases in the conditional correlations regarding market returns during the pandemic. This finding is supported by Okorie and Lin ( 2020 ), who suggest a fractal contagion effect of COVID‐19 on the stock market. They also highlight that this fractal contagion effect vanishes in the middle and long run for both stock market return and volatility. Likewise, He et al . ( 2020 ) argue that the impact of COVID‐19 on stock markets has bidirectional spillover effects between Asian countries and European and American countries. Nevertheless, there is no evidence to suggest that COVID‐19 has a negative impact on these countries' stock markets greater than the global average, as measured by the S&P Global 1200 index. In contrast, Tokic ( 2020 ) suggests that COVID‐19 will accelerate the trend of de‐globalisation and de‐dollarisation. Consistent with this finding, Zhang et al . ( 2020 ) suggest that countries respond differently to national‐level policies and the general development of the pandemic; specifically, they show that the US stock market has failed to take a leading role in this regard.

In view of the above discussion, the COVID‐19 outbreak has resulted in exaggerated fear, uncertainty and pressure on stock markets. Consistent with the EMH, market participants incorporate news about COVID‐19, especially the number of confirmed cases/deaths, into their valuation (Al‐Awadhi et al ., 2020 ; Ashraf, 2020 ; Erdem, 2020 ; Ramelli and Wagner, 2020 ). Nevertheless, the stock market seems to overreact to such news, resulting in stock market jumps and higher volatilities in the short run (Ashraf, 2020 ; Baker et al ., 2020 ; Okorie and Lin, 2020 ). There is some evidence to suggest that the spillover effect of COVID‐19 impacts global economies (Akhtaruzzaman et al ., 2020 ; He et al ., 2020 ; Okorie and Lin, 2020 ). Thus, we hypothesise:

H1: If COVID‐19 induces uncertainty in the stock market, then the increase in the number of COVID‐19 confirmed cases/deaths should increase volatility, jumps and co‐jumps while reducing stock returns.

The EMH suggests that competition among knowledgeable participants leads to a situation where stock prices incorporate all publicly available information. Consistent with this notion, research at the firm level suggests that the stock market reacts mostly to firms' pre‐existing conditions that affect their ability to endure the crisis. Firms with less leverage (Ramelli and Wagner, 2020 ), more cash holdings (Alfaro et al ., 2020 ; Ding et al ., 2020 ) and greater financial flexibility (De Vito and Gómez, 2020 ; Fahlenbrach et al ., 2020 ) experienced less negative stock returns during the COVID‐19 pandemic. Similarly, firms with better corporate social performance, as measured by environmental and social (ES) ratings, could suffer a lower decline in performance during a pandemic (Albuquerque et al ., 2020 ).

Other researchers explore how aggregate stock market returns across the world are responding to the COVID‐19 pandemic. For instance, Liu et al . ( 2020 ) examine the short‐term impact of the coronavirus outbreak on 21 leading stock market indices using an event study approach, finding that the COVID‐19 outbreak has adverse impacts on stock indices' abnormal returns. In addition, their panel fixed‐effect regression results suggest that COVID‐19 increases stock investors' fear and creates pessimistic sentiment regarding future returns. Gormsen and Koijen ( 2020 ) analyse investors' expectations about economic growth evolving across horizons in response to the pandemic and subsequent policy responses, revealing that the US fiscal stimulus (around 24 March 2020) boosted the stock market and long‐term growth but did little to increase short‐term growth expectations.

Previous studies also suggest that fiscal capacity shapes the degree to which countries can respond effectively to the pandemic and hence how stock markets respond. Countries whose fiscal response would be constrained by debt might be thought to be more vulnerable to a pandemic. In line with this notion, Ding et al . ( 2020 ) show that stock markets in richer economies, as measured by GDP per capita, have weathered the pandemic better than those in poorer economies. Gerding et al . ( 2020 ) also consider the relationship between corporate characteristics and stock price reactions. Using individual stock‐level data from more than 100 countries, they find that stock market responses were less negative in countries with higher fiscal capacity (i.e., lower debt‐to‐GDP ratios). Greppmair et al . ( 2020 ) suggest that during the COVID‐19 pandemic, short sellers have been trading on a combination of a firm's liquidity and a government's fiscal capacity. In addition, they find short‐selling activity to be focused on illiquid companies headquartered in countries with a low credit rating. However, some suggest that not all debt capacity variables impact the effectiveness of interventions and policies at curbing the pandemic. Zaremba et al . ( 2021a , 2021b ) show equity investors seem to factor only labour market conditions in the potential risks associated with the spread of the pandemic. They argue that unemployment has a negative impact on consumption, thus directly affecting the performance of the stock market. To reinvestigate the role of the debt capacity variable during the pandemic, we incorporate the debt capacity variables and control for unemployment. Economic risk denotes a country's ability to pay back its debts. A country with strong economic health should provide more reliable investment than a country with weaker finances. We thus propose the following:

H2: If a country with low economic risk 2 implies stronger fiscal capacity, then the country should experience less decline in stock indices and lower stock volatility and jumps during the pandemic.

Financial risk is also an important determinant of a country's fiscal capability. It is often defined as a country's ability to finance its trade debt obligations. Since a country's capability to generate foreign exchange directly affects the capacity to repay foreign debt, we expect that:

H3: If a country with low financial risk 3 implies stronger fiscal capacity, then the country should experience less decline in stock indices and lower stock volatility and jumps during the pandemic.

Previous studies suggest that national‐level political characteristics are important for crisis management and recovery (Bosancianu et al ., 2020 ; Greer et al ., 2020 ). In times of crisis the people turn to the state for leadership and unified action, and thus one may suppose that a country requires more political institutions with centralised power to take forceful action to control the spread of the pandemic (Zaremba et al ., 2021a , 2021b ). Consistent with this argument, Ding et al . ( 2020 ) find that a country with greater state power, relative to the power of individuals, experienced smaller stock price declines during the COVID‐19 pandemic. In contrast, Capelle‐Blancard and Desroziers ( 2020 ) show the country's legal origin appears to have had no influence on stock market responses in 74 countries from January to April 2020.

On the other hand, some may argue that legitimacy, credibility and the trust people have in government are necessary for the people to respond through collaborative engagement with public authorities to address crises (Bosancianu et al ., 2020 ; Greer et al ., 2020 ). Countries with greater press freedom can benefit from better information flow and public trust. This notion is in agreement with Painter and Qiu ( 2021 ) and Barrios and Hochberg ( 2020 ), who find that political beliefs determine the perception of risk associated with COVID‐19 and health‐related decisions. In a similar vein, using a panel regression analysis of 75 countries, Erdem ( 2020 ) shows that the adverse effects of COVID‐19 on the stock market are lower in freer countries. 4 Pástor and Veronesi ( 2013 ) examine the impact of political uncertainty on stock returns, identifying that political uncertainty causes serious panic in the stock market, especially when the economy is weak.

At the same time, the spread of the pandemic might reduce the political tensions in a country, as saving lives take precedence over threats posed by other groups. However, as time goes by, the pandemic may aggravate existing conflicts and trigger some forms of social disorder. This notion is consistent with that of Sharif et al . ( 2020 ), who document an unprecedented increase in geopolitical risk levels in the US driven by the COVID‐19 outbreak. 5

Overall, there is no clear pattern across countries regarding the relation between political characteristics and stock market responses. We hypothesise that a country with less political stability, i.e., high political risk, may potentially destabilise financial markets and exacerbate crises. Thus, we hypothesise:

H4: Countries with high political risk 6 experience higher stock volatility and jumps, and lower returns during the COVID‐19 pandemic.

Country risk is an important factor affecting the debt service capacity of borrowing countries. It often refers to the political, economic and financial risks that are unique to a specific country, and which might lead to unanticipated investment losses. In a broader sense, country risk is the degree to which political and economic unrest affect the securities of issuers doing business in a particular country. Prior research suggests that short sellers focus on less liquid companies headquartered in countries with a low credit rating (Greppmair et al ., 2020 ). Thus, we hypothesise:

H5: Countries with high country risk 7 experience higher stock volatility and jumps, and lower returns during the COVID‐19 pandemic.

Overall, it appears that stock markets integrated both new information about COVID‐19 and pre‐existing conditions that affected firms' ability to endure the crisis. Nevertheless, there is scant analysis of the impacts on the jumps of stock market returns. This paper attempts to provide the first empirical insights into the COVID‐19 pandemic and its effects on jumps and co‐jumps across countries.

3. Data and method

3.1. data and variables.

To construct our sample, we retrieve 5‐min intraday stock indices during the period 30 October 2019 to 13 May 2020 from Datascope provided by the Refinitiv database. The daily COVID‐19 data are from the European Centre for Disease Prevention and Control (ECDC). The ECDC reports the numbers of new COVID‐19 cases and deaths daily. The variables COVID and Death are the daily growth rates in the cumulative COVID‐19 confirmed cases and deaths, respectively.

Following the literature, such as Andersen et al . ( 2003 ), Chan et al . ( 2014 ), and Tanthanongsakkun et al . ( 2018 ), we utilise 5‐min interval returns to minimise the measurement error resulting from a decrease in microstructure biases. The return of the stock index is defined as the following:

where r t,j denotes the j th 5‐min return for a stock index during day t , M denotes the total number of 5‐min return intervals during any trading day, and R ( t ) defines the daily return on day t , derived from the 5‐min stock index.

The frequency of stock market index jumps during COVID‐19 could be considerably higher than other previous disease outbreaks (Baker et al ., 2020 ). To capture this unprecedented stock market reaction to COVID‐19, 8 we follow the analysis in Andersen et al . ( 2007 ) by decomposing the realised volatility into separate continuous and discontinuous (jump) components based on the bipower variation measures proposed by Barndorff‐Nielsen and Shephard ( 2004 , 2006 ) (see also Andersen et al ., 2010 , 2011 ; Chan et al ., 2014 ; Tanthanongsakkun et al ., 2018 ).

The volatility over the active part of the trading day t is measured by the quadratic variation

The first integrated variance term represents the contribution from the continuous price path, where N t gives the number of jumps over day t , and ∑ j = 0 N t k t , j 2 accounts for the corresponding contribution to the variance from the within‐day jumps. Hence, in the absence of jumps, the quadratic variation is simply the integrated volatility of the continuous sample path of the cumulative return process:

The components of Equation ( 2 ) are not directly observable. Instead, following prior literature, such as Andersen and Bollerslev ( 1998 ) and Andersen et al . ( 2003 ), non‐parametric daily realised volatility, RV ( t ), is defined using high‐frequency intra‐daily square returns as:

As suggested by Andersen and Bollerslev ( 1998 ) and Andersen et al . ( 2003 ), the realised volatility converges uniformly in probability to the quadratic variation process as the sampling frequency goes to infinity. That is, the realised volatility estimator does not consistently estimate integrated volatility as the measure captures both the continuous and discontinuous components of volatility. Thus, the biopower variation measures developed by Barndorff‐Nielsen and Shephard ( 2004 , 2006 ) are used to disentangle the two components of the quadratic variation process. In particular, they show that the bipower variation, BV ( t ), converges to the integrated volatility, IV (t), for M  → ∞:

Although the use of very high frequency financial price data could increase the precision of the biopower variation estimate, it can potentially be seriously contaminated by market microstructure noise. To diminish the effects of the local serial correlation induced by microstructure noise, Huang and Tauchen ( 2005 ) suggest using staggered observed returns in the biopower variation estimate:

where μ 1 = 2 π ≅ 0.79788 . The bipower variation measure defined above involves an additional stagger relative to the measure originally considered in Barndorff‐Nielsen and Shephard ( 2004 ), which makes it robust to certain types of market microstructure noise.

Combining the results in the previous equations, the difference between the realised variation and the bipower variation consistently estimates the jump contribution of the quadratic variation process, that is:

Following prior research, such as Huang and Tauchen ( 2005 ), Andersen et al . ( 2007 ) and Tanthanongsakkun et al . ( 2018 ), we consider small changes as measurement errors or part of the continuous sample path process and treat the large values of the changes as the significant jump component. To determine if a movement is a significant jump on day t , we compute the Z statistic as follows:

This follows an asymptotically standard normal distribution under the null hypothesis of no within‐day jumps, where:

Based on the significant jump detection test statistic, the realised measure of the jump contribution to the quadratic variation of the price process is then measured by:

where I (∙) denotes the indicator function and Ф α refers to the inverse of the standard normal distribution with a critical value of α .

Accordingly, we define integrated variance, CV ( t ), such that the non‐parametric measures for the jump and continuous components add up to realised volatility:

Clearly, the significant jump detection test requires a choice of α . Following prior studies, such as Andersen et al . ( 2010 ), Andersen et al . ( 2011 ) and Chan et al . ( 2014 ), we use a critical value of α  = 0.99.

It has previously been observed that the financial contagion follows a similar pattern to that of COVID‐19 (Akhtaruzzaman et al ., 2020 ; He et al ., 2020 ; Okorie and Lin, 2020 ). In addition, US markets were one of the main sources of a spillover effect to other markets (Syriopoulos et al ., 2015 ). To assess this pattern, we construct a co‐jump variable by summing the number of occurrences when both the stock index and S&P500 display significant jumps on a particular day.

Countries with greater economic development might be thought to be less susceptible to a pandemic (Ding et al ., 2020 ). Similarly, countries with a lower octogenarian population might also be less susceptible. To capture these potential effects, we include GDP and percentage of population aged above 65 ( Population ). The GDP and population data are from The World Bank for the year 2018. The GDP data are in current US dollars and are converted from domestic currencies using single year official exchange rates.

Country risk could be an important factor to explain the variation in stock markets across countries (Greer et al ., 2020 ; Greppmair et al ., 2020 ). We use country risk indices (composite risk rating index, political risk rating index, economic risk rating index, financial risk rating index, and unemployment risk rating) from Political Risk Services' International Country Risk Guide (ICRG) by the PRS Group.

According to ICRG, the composition risk index comprises 22 variables, representing three major components of country risk, namely economic, financial and political. There are five variables representing each of the economic and financial components of risk, whereas the political component is based on 12 variables. The economic risk rating measures a country's current economic strengths and weaknesses and reflects a country's ability to finance its official, commercial and trade debt obligations. Similarly, the financial risk rating reflects the ability and willingness of a country to service its trade and foreign debt obligations. Finally, the political risk rating measures the political stability of a country, which affects the country's ability to service its financial obligations. The political and the composite (financial and economic) risk indices are each based on 100 (50) points, and range from 0 to 100 (50). In all cases, the lower (higher) the risk points, the higher (lower) the associated risk. Thus, to allow a more intuitive interpretation, we define countries as having high risk factors if their risk rating points are within the first quartile of high‐risk factors, and construct Politic , Fin , Econ and Com dummy variables, representing political risk, financial risk, economic risk and the composition risk index, respectively. Each variable takes the value of 1 for countries with high‐risk factors and 0 otherwise.

Table  1 provides summary statistics regarding the cumulative number of COVID‐19 confirmed cases/death, daily growth rates and country risk indices in Panel A, and market quality in Panel B. Several counties have relatively high composite risk ratings (low risk), such as the United States, Italy and Spain. However, the United States has the highest number of confirmed cases and death. Italy and Spain, on the other hand, have the highest growth rates of confirmed cases and deaths, respectively. Yet, the market quality measures are all positive for the United States, while the market quality measures of several counties have mixed responses to the COVID‐19 pandemic information. It is therefore interesting to formally test the relation between market quality and severity of COVID‐19 given each country's risks such as economic, finance and political risks in the next section.

Summary statistics

Panel A reports the number and the growth of confirmed cases, deaths, and risk ratings (Composte, Economic, Finance and Political risk ratings) for each country. The average daily return, realised volatility, jumps and co‐jumps with S&P500 are reported in Panel B. Our sample encompasses 30 October 2019 to 13 May 2020.

3.2. Methodology

To examine the impact of changes in COVID‐19 confirmed cases/deaths on market quality, we use high‐frequency data on daily stock indices to obtain a measurement of market quality. The following baseline model is used:

where the dependent variable, Y , is market quality and proxied by the return, realised volatility, jumps and co‐jumps. Our key independent variables is COVID , which is either (i) daily growth in total confirmed cases, (ii) daily growth in total cases of deaths or (iii) both daily growth in total confirmed cases and deaths caused by COVID‐19. Control comprises control variables, such as GDP, population, unemployment, one period lag growth rate in the cumulative number of confirmed COVID‐19 cases/deaths, one period lag stock return, and one period lag realised volatility.

Next, to understand how the country risk and its components (i.e., economic, political and financial) influence the relation between COVID‐19 and market quality, we repeat our analyses with additional variables capturing different aspects of country risk. These include Econ , Politic , Fin and Com , representing economic risk, political risk, financial risk and the composition risk index, respectively. We define Econ as a dummy variable that is equal to one if the country has high economic risk, and zero otherwise; Politic as a dummy variable that is equal to one if the country has high political risk, and zero otherwise; Fin as a dummy variable that is equal to one if the country has high financial risk, and zero otherwise; and Com as a dummy variable that is equal to one if the country has high composite risk, and zero otherwise. Finally, we also include the interaction terms between these risks and the growth in the number of confirmed cases and deaths. To explore the impact of the country risk on the relationship between COVID‐19 and the stock market, we run the following regression:

where the dependent variable, Y , is market quality and proxied by the return, realised volatility, jumps and co‐jumps. COVID is as defined for Equation ( 12 ). The key explanatory variables are RISK and its interactions with COVID . RISK represents either Econ , Politic , Fin or Com . Control comprises the same variables as for Equation ( 12 ).

4. Empirical results

Table  2 reports the baseline regression results of panel data for 43 stock indices around the world. The results suggest that COVID‐19 (i.e., the growth in the number of confirmed cases) has a positive and a significant impact on financial volatility, jumps and co‐jumps, but a negative impact on financial returns. This finding is in line with previous studies that also identify the adverse effect of COVID‐19 on stock market quality (Alan et al ., 2020 ; Ashraf, 2020 ; Baker et al ., 2020 ; Ramelli and Wagner, 2020 ). This finding implies that, during the COVID‐19 pandemic, market participants incorporate news about the pandemic into their valuation. Another possible explanation for this finding is that the COVID‐19 pandemic changed the way market participants perceive risk, which results in an increased volatility of markets due to more homogeneous beliefs of market participants who expect higher levels of risk (Burns et al ., 2012 ). Furthermore, the coefficient of COVID in the co‐jumps model is positive and significant, suggesting a possible spillover effect of the pandemic. When the number of confirmed cases increases, stock market indices around the world appear to jump with the US stock market. These results are consistent with findings of spillover effects between Asian countries and European and American countries (Akhtaruzzaman et al ., 2020 ; He et al ., 2020 ; Okorie and Lin, 2020 ). In contrast, the growth rate of cumulative deaths only has a positive impact on the realised volatility model. This result may be explained by the fact that the market participants are already pricing the effect of the pandemic by using new confirmed cases (Ashraf, 2020 ). To ensure that our results are not driven by multicollinearity between the daily growth rate in confirmed cases and the daily growth rate in deaths, we also run two separate regressions with each of these two variables representing COVID‐19 infection. 9 The results of growth in death/cases remain similar. Furthermore, our regressions include one period lag in the growth rate of the cumulative number of confirmed COVID‐19 cases/deaths, which capture the impact of past confirmed cases/death growth rate on the current stock market performance. Thus, our results remain similar and are unlikely driven by historical growth rate of COVID‐19 cases/deaths.

Baseline panel regression of the relation between COVID‐19 cases/deaths data and market quality

This table reports results from our baseline panel regression, where dependent variables are daily return (Return), daily realised volatility (RV), and jumps and co‐jumps with S&P500. Our key variables of interest are the growth rate of cumulative confirmed cases ( COVID ) and the growth rate of cumulative death cases ( Death ). We also control for the percentage of the population aged above 65, GDP and unemployment risk. For return and realised volatility, we also control for lagged return and lagged RV. Our sample encompasses 30 October 2019 to 13 May 2020. The robust standard errors are reported in parentheses. ***, **, * indicates significance at the 1, 5 and 10 percent levels, respectively.

Table  3 reports the effect of economic risk on the relation between the growth in COVID‐19 confirmed cases/deaths and market quality. Consistent with the baseline model, COVID‐19 has a significant impact on market quality. Surprisingly, in all the models, we fail to highlight the impact of economic risk on market quality when there is an exogenous economic shock from the pandemic. This finding is contrary to previous studies that examine the impact of the pandemic at the firm level, which suggests that stock markets in richer economies suffer less during the crisis (Ding et al ., 2020 ). This inconsistent finding could be due to different samples, periods of study, level of analysis and control variables. Unlike previous studies, we run the analysis at the aggregate country level. In addition, we also control for the unemployment factor, which appears to significantly influence the country‐level financial immunity to the pandemic (Zaremba et al ., 2021a , 2021b ). Though the economic risk may influence the country's ability to pay back its debts, it indirectly affects the performance of the stock market. For this reason, this information may not be priced in by stock market investors.

Effect of economic risk on the relation between COVID‐19 cases/deaths data and market quality

This table reports results from our panel regression, where dependent variables are daily return (Return), daily realised volatility (RV), and jumps and co‐jumps with S&P500. Our key variables of interest are the growth rate of cumulative confirmed cases ( COVID ) and the growth rate of cumulative death cases ( Death ). We also control for the percentage of the population aged above 65, GDP and unemployment risk. For return and realised volatility, we also control for lagged return and lagged RV. Our sample encompasses 30 October 2019 to 13 May 2020. The robust standard errors are reported in parentheses. ***, **, * indicates significance at the 1, 5 and 10 percent levels, respectively. Econ is a dummy variable equal to one for a country with a high economic risk rating, zero otherwise.

Previous studies suggest that country‐level political characteristics can play a role in explaining the stock market reaction to COVID‐19 (Bosancianu et al ., 2020 ; Ding et al ., 2020 ; Erdem, 2020 ; Greer et al ., 2020 ). In line with these studies, we repeat our analyses considering political risk (Table  4 ); the coefficients for COVID‐19 confirmed cases remain significant in all models, while confirmed death is only significant in the realised volatility models. This result is consistent with our main finding. Our focus, however, is on the interaction term between COVID and political risk. This interaction term shows the marginal effect of COVID on market quality when a country has high political risk. The regression analysis in Table  4 shows the interaction term for the jump model is negative and statistically significant. 10 This suggests that countries with low political stability experienced lower volatility in stock indices as the number of COVID‐19 cases grew. A possible explanation for this might be that during the pandemic, people turned to the state for leadership and unified action, and thus countries with centralised power are likely to have taken forceful or appropriate action to prevent the spread of the virus, resulting in less panic in the stock market. This finding is supported by Ding et al . ( 2020 ), who find that countries with civil and socialist legal traditions experienced less decline in stock prices than those with a common law tradition. Along similar lines, Zaremba et al . ( 2021a , 2021b ) points out that countries with less freedom of expression were better able to cope with the adverse consequences of the pandemic.

Effect of political risk on the relation between COVID‐19 cases/deaths data and market quality

This table reports results from our panel regression, where dependent variables are daily return (Return), daily realised volatility (RV), and jumps and co‐jumps with S&P500. Our key variables of interest are the growth rate of cumulative confirmed cases ( COVID ) and the growth rate of cumulative death cases ( Death ). We also control for the percentage of the population aged above 65, GDP and unemployment risk. For return and realised volatility, we also control for lagged return and lagged RV. Our sample encompasses 30 October 2019 to 13 May 2020. The robust standard errors are reported in parentheses. ***, **, * indicates significance at the 1, 5 and 10 percent levels, respectively. Politic is a dummy variable equal to one for a country with a high political risk rating, zero otherwise.

Table  5 focuses on the impact of fiscal capacity on stock market returns during the COVID‐19 crisis. We find that the coefficients for the interaction terms between COVID and financial risk are insignificant in the full models. However, when we examine only the confirmed COVID‐19 cases, the interaction term is negative and significant in the return model. 9 This implies that countries with high financial risk were able to ameliorate the adverse effects of COVID‐19 on market returns. These results are consistent with other studies (Gerding et al ., 2020 ; Greppmair et al ., 2020 ). This result may be explained by the fact that countries with greater financial flexibility are more able to fund an appropriate stimulus package, which is used to offset the effects of the pandemic.

Effect of financial risk on the relation between COVID‐19 cases/deaths data and market quality

This table reports results from our panel regression, where dependent variables are daily return (Return), daily realised volatility (RV), and jumps and co‐jumps with S&P500. Our key variables of interest are the growth rate of cumulative confirmed cases ( COVID ) and the growth rate of cumulative death cases ( Death ). We also control for the percentage of the population aged above 65, GDP and unemployment risk. For return and realised volatility, we also control for lagged return and lagged RV. Our sample encompasses 30 October 2019 to 13 May 2020. The robust standard errors are reported in parentheses. ***, **, * indicates significance at the 1, 5 and 10 percent levels, respectively. Fin is a dummy variable equal to one for a country with a high financial risk rating, zero otherwise.

To evaluate how country risk shapes stock price movements in response to the COVID‐19 pandemic, we retest our baseline model by using composite risk as a proxy for country risk (Table  6 ). The composite risk is a simple function of the economic, political and financial risk indices. Consistent with our baseline model, the coefficient of COVID is negative and significant for market return and positively related to realised volatility, jumps and co‐jumps of the stock indices. The interaction terms between the COVID‐19 confirmed cases, deaths, and the composite risk index are negative and significant for the realised volatility models. 11 This may suggest that countries with low stability overall experience lower volatility in their stock indices. Although this result is rather surprising, one explanation is that in countries with low stability, people often must rely on themselves and react to the pandemic sooner, thus resulting in a lower volatility in the markets. This is also consistent with Abuzayed et al . ( 2021 ) who find that developed markets transmitted and received more marginal extreme risk during the COVID‐19 pandemic. Unfortunately, for other measures of market quality, this study finds a weak association with country risk. Thus, it is not clear whether stock markets in richer economies, more indebted countries or with more state power have reacted differently to COVID‐19. A possible explanation for this result is that economic and political risks can be intertwined. For instance, a country with strong economic health may not be a good candidate for investment if the political climate is unwelcoming to outside investors.

Effect of composite risk on the relation between COVID‐19 cases/deaths data and market quality

This table reports results from our panel regression, where dependent variables are daily return (Return), daily realised volatility (RV), and jumps and co‐jumps with S&P500. Our key variables of interest are the growth rate of cumulative confirmed cases ( COVID ) and the growth rate of cumulative death cases ( Death ). We also control for the percentage of the population aged above 65, GDP and unemployment risk. For return and realised volatility, we also control for lagged return and lagged RV. Our sample encompasses 30 October 2019 to 13 May 2020. The robust standard errors are reported in parentheses. ***, **, * indicates significance at the 1, 5 and 10 percent levels, respectively. Com is a dummy variable equal to one for a country with a high composite risk rating, zero otherwise.

5. Conclusion

We have examined the impact of COVID‐19 on financial markets around the world by utilising intraday data and the Barndorff‐Nielsen and Shephard (2004) nonparametric jump detection technique. To this end, we have used stock return, realised volatility, jumps and co‐jumps as a proxy for market quality and we have explored whether country risk plays a significant role in the relation between COVID‐19 and market quality. The outcomes of our empirical investigation underline the fact that: (i) the growth in cumulative COVID‐19 confirmed cases amplifies realised volatility and jumps while reducing returns; (ii) the impact of COVID‐19 on volatility is weaker in high political risk countries; and (iii) the impact of COVID‐19 on market return is stronger in high financial risk countries. Our findings have important implications for financial market participants. This study provides insights about the stock market response to the pandemic and how country characteristics play an important role in shaping the stock market response to COVID‐19‐induced financial market instability. Future research could potentially evaluate different jump detection techniques for extremely volatile periods similar to the COVID‐19 pandemic.

We would like to thank Professor Jing Shi (Editor), and an anonymous reviewer for their valuable comments and suggestions. This research was funded by Chulalongkorn University under the Ratchadapisek Sompoch Endowment Fund through the Center of Excellence (CE) in Management Research for Corporate Governance and Behavioral Finance and Sasin School of Management through SASIN Major Grant for a research program.

1 For instance, Ashraf ( 2020 ) suggests that stock markets react strongly with negative returns to growth in confirmed cases; however, response to the growth in deaths is not statistically significant. Al‐Awadhi et al . ( 2020 ) and Erdem ( 2020 ) indicate that both the daily growth in total confirmed cases and in total deaths caused by COVID‐19 have significant negative effects on stock returns.

2 We use the economic risk index from Political Risk Services' International Country Risk Guide (ICRG) by the PRS Group. It reflects a country's ability to finance its official, commercial and trade debt obligations by using five variables, namely GDP per head, real GDP growth, annual inflation rate, budget balance as a percentage of GDP, and current account as a percentage of GDP.

3 We use the financial risk index from ICRG by the PRS Group. It reflects a country's ability to finance through inflows of foreign exchange by using five variables, namely foreign debt as a percentage of GDP, foreign debt service as a percentage of exports in goods and services, current account as a percentage of exports in goods and services, net liquidity as months of import cover, and exchange rate stability.

4 Erdem ( 2020 ) uses the Human Freedom Index 2019 from Freedom House as a proxy for the level of a country's freedom. This index adds scores of 10 political rights indicators and 15 civil liberties indicators.

5 Sharif et al . ( 2020 ) use the GPR index as a proxy for geopolitical risk. This index is constructed based on news related to geopolitical events. The number of words related to geopolitical risk are counted each day in each newspaper to calculate the daily GPR index.

6 We use the political risk index from ICRG by the PRS Group. It reflects a country’s political stability by using 12 variables, namely government stability, socio‐economic conditions, investment profile, internal conflict, external conflict, corruption, military in politics, religious tension, law and order, ethnic tensions, democratic accountability, and bureaucracy quality.

7 The country risk index reflects the uncertainty associated with investing in a particular country. It comprises 22 variables, representing three major components of country risk, namely economic, financial and political.

8 Identifying a jump is simply a way to construct one of our dependent variables. The nonparametric jump technique dates back to 2002 and four approaches are popular in the existing literature. We choose the Barndorff‐Nielsen and Shephard ( 2004 , 2006 ) approach as it is one of the earliest approaches and has been successfully adopted by various authors from 2002 until recently in 2021 (see Andersen et al ., 2003 ; Tanthanongsakkun et al ., 2018 ; Ho et al ., 2021 ; Phiromswad et al ., 2021 ).

9 We run separate regressions for the daily growth rate in confirmed cases and the daily growth rate in deaths in all analyses. The results remain consistent. To save space, these results are not reported and are available upon request.

10 When we examine the COVID‐19 confirmed cases only, the interaction terms in the RV and jump models are also negative and statistically significant. To save space, the results are not reported here, but are available upon request.

11 For realised volatility, the interaction terms between the COVID‐19 confirmed cases, death, and the composite risk in the separate models are also negative and significant. The results are not reported here and are available upon request.

• Abuzayed, B., Bouri E., Al‐Fayoumi N., and Jalkh N., 2021, Systemic risk spillover across global and country stock markets during the COVID‐19 pandemic , Economic Analysis and Policy 71 , 180–197. [ Google Scholar ]
• Akhtaruzzaman, M., Boubaker S., and Sensoy A., 2020, Financial contagion during COVID–19 crisis , Finance Research Letters 38 , 101604. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Alan, N. S., Engle R. F., and Karagozoglu A. K., 2020, Multi‐regime Forecasting Model for the impact of COVID‐19 pandemic on volatility in global equity markets . 10.2139/ssrn.3646520 [ CrossRef ]
• Al‐Awadhi, A. M., Alsaifi K., Al‐Awadhi A., and Alhammadi S., 2020, Death and contagious infectious diseases: impact of the COVID‐19 virus on stock market returns , Journal of Behavioral and Experimental Finance 27 , 100326. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Albuquerque, R., Koskinen Y., Yang S., and Zhang C., 2020, Resiliency of environmental and social stocks: an analysis of the exogenous COVID‐19 market crash , The Review of Corporate Finance Studies 9 , 593–621. [ Google Scholar ]
• Alfaro, L., Chari A., Greenland A. N. & Schott P. K., 2020, Aggregate and firm‐level stock returns during pandemics, in real time, NBER Working Paper Series No. 26950 . Available at: http://www.nber.org/papers/w26950 .
• Altig, D., Baker S., Barrero J. M. et al ., 2020, Economic uncertainty before and during the COVID‐19 pandemic , Journal of Public Economics 191 , 104274. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Andersen, T. G., and Bollerslev T., 1998, Deutsche mark‐dollar volatility: intraday activity patterns, macroeconomic announcements, and longer run dependencies , The Journal of Finance 53 , 219–265. [ Google Scholar ]
• Andersen, T. G., Bollerslev T., and Diebold F. X., 2007, Roughing it up: including jump components in the measurement, modeling, and forecasting of return volatility , The Review of Economics and Statistics 89 , 701–720. [ Google Scholar ]
• Andersen, T. G., Bollerslev T., Diebold F. X., and Ebens H., 2001, The distribution of realized stock return volatility , Journal of Financial Economics 61 , 43–76. [ Google Scholar ]
• Andersen, T. G., Bollerslev T., Diebold F. X., and Labys P., 2003, Modeling and forecasting realized volatility , Econometrica 71 , 579–625. [ Google Scholar ]
• Andersen, T. G., Bollerslev T., Frederiksen P., and Ørregaard Nielsen M., 2010, Continuous‐time models, realized volatilities, and testable distributional implications for daily stock returns , Journal of Applied Econometrics 25 , 233–261. [ Google Scholar ]
• Andersen, T. G., Bollerslev T., and Huang X., 2011, A reduced form framework for modeling volatility of speculative prices based on realized variation measures , Journal of Econometrics 160 , 176–189. [ Google Scholar ]
• Ashraf, B. N., 2020, Stock markets’ reaction to COVID‐19: cases or fatalities? Research in International Business and Finance 54 , 101249. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Baker, S. R., Bloom N., Davis S. J., Kost K., Sammon M., and Viratyosin T., 2020, The unprecedented stock market reaction to COVID‐19 , The Review of Asset Pricing Studies 10 , 742–758. [ Google Scholar ]
• Baldwin, R., and Di Mauro B. W., 2020, Economics in the time of COVID‐19: a new eBook (VOX CEPR Policy Portal, London; ). [ Google Scholar ]
• Barndorff‐Nielsen, O. E., and Shephard N., 2004, Power and bipower variation with stochastic volatility and jumps , Journal of Financial Econometrics 2 , 1–37. [ Google Scholar ]
• Barndorff‐Nielsen, O. E., and Shephard N., 2006, Econometrics of testing for jumps in financial economics using bipower variation , Journal of Financial Econometrics 4 , 1–30. [ Google Scholar ]
• Barrios, J. M. & Hochberg Y. 2020, Risk perception through the lens of politics in the time of the Covid‐19 pandemic, NBER Working Paper Series No. 27008. Available at: http://www.nber.org/papers/w27008.
• Black, F., 1986, Noise , The Journal of Finance 41 , 528–543. [ Google Scholar ]
• Bosancianu, C. M., Dionne K. Y., Hilbig H. et al ., 2020, Political and social correlates of Covid‐19 mortality , SocArXiv. 10.31235/osf.io/ub3zd. [ CrossRef ]
• Burns, W. J., Peters E., and Slovic P., 2012, Risk perception and the economic crisis: a longitudinal study of the trajectory of perceived risk , Risk Analysis 32 , 659–677. [ PubMed ] [ Google Scholar ]
• Capelle‐Blancard, G., and Desroziers A., 2020, The stock market is not the economy? Insights from the COVID‐19 crisis , CEPR Covid Economics. 10.2139/ssrn.3638208. [ CrossRef ]
• Chan, K. F., Powell J. G., and Treepongkaruna S., 2014, Currency jumps and crises: do developed and emerging market currencies jump together? Pacific‐Basin Finance Journal 30 , 132–157. [ Google Scholar ]
• Cheng, I.‐H., 2020, Volatility markets underreacted to the early stages of the COVID‐19 pandemic , The Review of Asset Pricing Studies 10 , 635–668. [ Google Scholar ]
• De Long, J. B., Shleifer A., Summers L. H., and Waldmann R. J., 1990, Noise trader risk in financial markets , Journal of Political Economy 98 , 703–738. [ Google Scholar ]
• De Vito, A., and Gómez J.‐P., 2020, Estimating the COVID‐19 cash crunch: global evidence and policy , Journal of Accounting and Public Policy 39 , 106741. [ Google Scholar ]
• Ding, W., Levine R., Lin C. & Xie W. 2020, Corporate immunity to the COVID‐19 pandemic, NBER Working Paper Series No. 27055. Available at: http://www.nber.org/papers/w27055.
• Dungey, M., Matei M., and Treepongkaruna S., 2014, Identifying periods of financial stress in Asian currencies: the role of high frequency financial market data , report (University of Tasmania). Available at: https://eprints.utas.edu.au/18605/ .
• Erdem, O., 2020, Freedom and stock market performance during Covid‐19 outbreak , Finance Research Letters 36 , 101671. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Fahlenbrach, R., Rageth K., and Stulz R. M., 2020, How valuable is financial flexibility when revenue stops? Evidence from the Covid‐19 crisis , NBER Working Paper Series No. 27106. Available at: http://www.nber.org/papers/w27106 .
• Gerding, F., Martin T., and Nagler F., 2020, Sovereign debt and equity returns in the face of disaster . 10.2139/ssrn.3572839. [ CrossRef ]
• Goodell, J. W., 2020, COVID‐19 and finance: agendas for future research , Finance Research Letters 35 , 101512. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Gormsen, N. J., and Koijen R. S. J., 2020, Coronavirus: impact on stock prices and growth expectations , The Review of Asset Pricing Studies 10 , 574–597. [ Google Scholar ]
• Greer, S. L., King E. J., da Fonseca E. M., and Peralta‐Santos A., 2020, The comparative politics of COVID‐19: the need to understand government responses , Global Public Health 15 , 1413–1416. [ PubMed ] [ Google Scholar ]
• Greppmair, S., Jank S., and Smajlbegovic E., 2020, Betting on disaster: short‐selling activity during the COVID‐19 pandemic . 10.2139/ssrn.3638584. [ CrossRef ]
• He, Q., Liu J., Wang S., and Yu J., 2020, The impact of COVID‐19 on stock markets , Economic and Political Studies 8 , 275–288. [ Google Scholar ]
• Ho, G. K. F., Treepongkaruna S., Wee M., and Padungsaksawasdi C., 2021, The effect of short selling on volatility and jumps , Australian Journal of Management . 10.1177/0312896221996416. [ CrossRef ] [ Google Scholar ]
• Huang, X., and Tauchen G., 2005, The relative contribution of jumps to total price variance , Journal of Financial Econometrics 3 , 456–499. [ Google Scholar ]
• Jackwerth, J., 2020, What do index options teach us about COVID‐19? The Review of Asset Pricing Studies 10 , 618–634. [ Google Scholar ]
• Lee, W. Y., Jiang C. X., and Indro D. C., 2002, Stock market volatility, excess returns, and the role of investor sentiment , Journal of Banking and Finance 26 , 2277–2299. [ Google Scholar ]
• Liu, H. Y., Manzoor A., Wang C. Y., Zhang L., and Manzoor Z., 2020, The COVID‐19 outbreak and affected countries stock markets response , International Journal of Environmental Research and Public Health 17 , 2800. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Mirza, N., Naqvi B., Rahat B., and Rizvi S. K. A., 2020, Price reaction, volatility timing and funds' performance during Covid‐19 , Finance Research Letters 36 , 101657. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Okorie, D. I., and Lin B., 2020, Stock markets and the COVID‐19 fractal contagion effects , Finance Research Letters 38 , 101640. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Painter, M., and Qiu T., 2021, Political beliefs affect compliance with Covid‐19 social distancing orders , Journal of Economic Behavior and Organization 185 , 688–701. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Pástor, Ľ., and Veronesi P., 2013, Political uncertainty and risk premia , Journal of Financial Economics 110 , 520–545. [ Google Scholar ]
• Phiromswad, P., Chatjuthamard P., Treepongkaruna S., and Srivannaboon S., 2021, Jumps and cojumps analyses of major and minor cryptocurrencies , PLoS One 16 , e0245744. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Ramelli, S., and Wagner A. F., 2020, Feverish stock price reactions to COVID‐19 , The Review of Corporate Finance Studies 9 , 622–655. [ Google Scholar ]
• Roszkowski, M. J., and Davey G., 2010, Risk perception and risk tolerance changes attributable to the 2008 economic crisis: a subtle but critical difference , Journal of Financial Service Professionals 64 , 42–53. [ Google Scholar ]
• Sharif, A., Aloui C., and Yarovaya L., 2020, COVID‐19 pandemic, oil prices, stock market, geopolitical risk and policy uncertainty nexus in the US economy: fresh evidence from the wavelet‐based approach , International Review of Financial Analysis 70 , 101496. [ Google Scholar ]
• Syriopoulos, T., Makram B., and Boubaker A., 2015, Stock market volatility spillovers and portfolio hedging: BRICS and the financial crisis , International Review of Financial Analysis 39 , 7–18. [ Google Scholar ]
• Tanthanongsakkun, S., Treepongkaruna S., Wee M., and Brooks R., 2018, The effect of trading by different trader types on realized volatility and jumps: evidence from the Thai stock market , Chulalongkorn Business Review 40 , 111–142. [ Google Scholar ]
• Tokic, D., 2020, Long‐term consequences of the 2020 coronavirus pandemics: historical global‐macro context , Journal of Corporate Accounting and Finance 31 , 9–14. [ Google Scholar ]
• Wang, J., Lu X., He F., and Ma F., 2020, Which popular predictor is more useful to forecast international stock markets during the coronavirus pandemic: VIX vs EPU? International Review of Financial Analysis 72 , 101596. [ Google Scholar ]
• Zaremba, A., Aharon D. Y., Demir E., Kizys R., and Zawadka D., 2021a, COVID‐19, government policy responses, and stock market liquidity around the world: a note , Research in International Business and Finance 56 , 101359. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Zaremba, A., Kizys R., Aharon D. Y., and Demir E., 2020, Infected markets: novel coronavirus, government interventions, and stock return volatility around the globe , Finance Research Letters 35 , 101597. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Zaremba, A., Kizys R., Tzouvanas P., Aharon D. Y., and Demir E., 2021b, The quest for multidimensional financial immunity to the COVID‐19 pandemic: evidence from international stock markets , Journal of International Financial Markets, Institutions and Money 71 , 101284. [ Google Scholar ]
• Zhang, D., Hu M., and Ji Q., 2020, Financial markets under the global pandemic of COVID‐19 , Finance Research Letters 36 , 101528. [ PMC free article ] [ PubMed ] [ Google Scholar ]

Advertisement

Supported by

‘It’s Clearly Bleak’: Stocks Notch Longest Losing Streak in Months

A rally at the start of the year has given way to worries on Wall Street about economics and geopolitics.

• Share full article

S&P 500

By Joe Rennison

Stocks suffered their longest losing streak of the year, as geopolitical turmoil rattled Wall Street and investors slashed their bets on the Federal Reserve cutting interest rates any time soon.

The S&P 500 fell 0.9 percent on Friday, its sixth consecutive decline, marking its worst run since October 2022.

The slide dragged the S&P 500 down by just over 3 percent for the week, a third straight weekly decline. By that measure, it is the longest weekly losing streak for the index since September, when concerns over rising government debt and a potential government shutdown compounded worries about the effects of high interest rates.

Those fears dissipated toward the end of last year as inflation cooled and investors began to bet that the Fed would soon cut rates, prompting a ferocious stock rally in the first three months of 2024.

But this month, worries that stubborn inflation would lead the Fed to keep rates high have returned, compounded by the widening conflict in the Middle East, with Israel striking Iran early on Friday .

“It’s clearly bleak,” said Andrew Brenner, head of international fixed income at National Alliance Securities.

Investors have pulled roughly $21 billion out of funds that invest in U.S. stocks over the two weeks through Wednesday, according to data from EPFR Global, which tracks fund flows. That compares to an inflow of around $80 billion for the year through early April. And the unease is not just apparent in the stock market.

U.S. government bond yields, which underpin interest rates for a wide variety of loans, have been rising. The average rate on 30-year mortgages, the most popular home loan in the United States, rose above 7 percent on Thursday for the first time this year.

The dollar is also markedly higher, putting pressure on countries that import goods from the United States and issue dollar-denominated debt. And oil prices, stoked by geopolitical tensions, are up more than 13 percent since the start of the year.

“There is nothing that looks good right now,” Mr. Brenner said.

Recent reports showing hotter-than-expected inflation have altered investors’ forecasts for the Fed , which has kept its key rate near a two-decade high. “The recent data have clearly not given us greater confidence and instead indicate that it’s likely to take longer than expected to achieve that confidence,” Jerome H. Powell, the Fed chair, said at an event in Washington on Tuesday.

Economists at Société Générale no longer expect the Fed to cut rates this year. BNP Paribas and Wells Fargo economists have also dialed down their expectations for cuts.

Traders in futures markets, which allow investors to bet on where interest rates are headed, are wagering on one, and perhaps two, quarter-point cuts by the end of the year. At the start of the year, traders were expecting six cuts over that period.

At first, the shift appeared to be welcomed by stock investors. A strong economy, all else equal, is good for the stock market, and while some inflation data had started to buck the trend earlier this year it wasn’t enough to disrupt the broader cooling that took hold in 2023. But recent inflation reports have continued to disappointed investors and economists and become harder to ignore.

John Williams, the president of the New York Fed, said this week that it was possible that another increase, rather than a cut, to rates might be warranted if inflation remained sticky, even if that wasn’t what he considered the most likely scenario. Other officials have noted that the Fed may have to wait until much later this year, or even 2025, to begin easing rates.

So far, worries have yet to intensify to the point of threatening the strength of the U.S. economy. Although the S&P 500 has fallen 5.5 percent this month, it remains more than 4 percent higher for the year.

And a recent survey of fund managers around the world by Bank of America showed the most optimism since January 2022, with respondents expecting global growth to accelerate. The biggest risk, according to the respondents, is a rise in inflation that could keep interest rates elevated, squeezing growth abroad and at home.

Reflecting some of those worries, the Russell 2000 index that tracks smaller companies that tend to be more sensitive to the outlook for the domestic economy, has slumped to a loss of roughly 4 percent for the year.

“I think the recent selling pressure is just the beginning of a larger move,” said Peter Tchir, head of macro strategy at Academy Securities.

An earlier version of this article misstated the number of consecutive weeks the S&P 500 is set to drop. The market moves this week have set the stock index up for a third straight weekly decline, not a fourth.

How we handle corrections

Joe Rennison writes about financial markets, a beat that ranges from chronicling the vagaries of the stock market to explaining the often-inscrutable trading decisions of Wall Street insiders. More about Joe Rennison

The stock market can rise another 13% this year even with a dwindling rate-cut outlook, Fundstrat's Tom Lee says

• Stocks could still perform well this year even if rate cuts are fewer than markets initially expected. 
• Fundstrat's Tom Lee predicted the S&P 500 could rise to 5,700 by the end of the year.
• Equities will be powered higher by a strong economy and generally cooling inflation, he said.

Dwindling hopes for Fed rate cuts won't necessarily derail the stock market's upward trajectory through the rest of this year, according to Fundstrat's head of research Tom Lee.

Lee, one of the most bullish forecasters on Wall Street, predicted the S&P 500 could jump to 5,700 by the end of the year, implying another 13% upside for the benchmark index. The market doesn't need Fed rate cuts to do well, he said in a recent interview with CNBC , assuming that the economy remains strong and inflation continues to cool.

The economy looks like it's meeting those conditions so far, Lee added. Corporate earnings look strong, with the S&P 500 on track to notch at least 7% earnings growth this quarter, according to FactSet . Economic growth has also remained resilient, with the economy expected to grow 2.9% over the first quarter, per estimates from the Atlanta Fed.

And while headline inflation came in hotter than expected in March, most components of the consumer price index are actually posting around 2% year-per-year price growth, in-line with the Fed's long-run target. Not accounting for auto, housing, energy, and food prices, annualized inflation clocked in at 2.7% last month, Lee said, suggesting that inflation was cooling overall.

"We don't really need the Fed to make three cuts," he added.

But the only risk that looms over stocks is if inflation comes in hotter than expected, prompting the Fed to issue another rate hike, Lee warned. 

"I think that's still very much a tail scenario, but that would be the one to rattle markets the most," he said. 

Hot inflation figures have led investors to dial back their outlooks for Fed rate cuts this year, with stocks sliding over the past week as markets repriced their expectations. Investors are now expecting just one or two cuts in 2024, according to the CME FedWatch tool, down from six cuts priced in earlier this year.

• Main content

Numbers, Facts and Trends Shaping Your World

Read our research on:

Full Topic List

Regions & Countries

• Publications
• Our Methods
• Short Reads
• Tools & Resources

Read Our Research On:

Majority of workers who quit a job in 2021 cite low pay, no opportunities for advancement, feeling disrespected

The COVID-19 pandemic set off nearly unprecedented churn in the U.S. labor market. Widespread job losses in the early months of the pandemic gave way to tight labor markets in 2021, driven in part by what’s come to be known as the Great Resignation . The nation’s “quit rate” reached a 20-year high last November.

A new Pew Research Center survey finds that low pay, a lack of opportunities for advancement and feeling disrespected at work are the top reasons why Americans quit their jobs last year. The survey also finds that those who quit and are now employed elsewhere are more likely than not to say their current job has better pay, more opportunities for advancement and more work-life balance and flexibility.

Majorities of workers who quit a job in 2021 say low pay (63%), no opportunities for advancement (63%) and feeling disrespected at work (57%) were reasons why they quit, according to the Feb. 7-13 survey. At least a third say each of these were major reasons why they left.  

Roughly half say child care issues were a reason they quit a job (48% among those with a child younger than 18 in the household). A similar share point to a lack of flexibility to choose when they put in their hours (45%) or not having good benefits such as health insurance and paid time off (43%). Roughly a quarter say each of these was a major reason.

Pew Research Center conducted this analysis to better understand the experiences of Americans who quit a job in 2021. This analysis is based on 6,627 non-retired U.S. adults, including 965 who say they left a job by choice last year. The data was collected as a part of a larger survey conducted Feb. 7-13, 2022. Everyone who took part is a member of the Center’s American Trends Panel (ATP), an online survey panel that is recruited through national, random sampling of residential addresses. This way, nearly all U.S. adults have a chance of selection. The survey is weighted to be representative of the U.S. adult population by gender, race, ethnicity, partisan affiliation, education and other categories. Read more about the ATP’s methodology .

Here are the questions used for this analysis, along with responses, and its methodology.

About four-in-ten adults who quit a job last year (39%) say a reason was that they were working too many hours, while three-in-ten cite working too few hours. About a third (35%) cite wanting to relocate to a different area, while relatively few (18%) cite their employer requiring a COVID-19 vaccine as a reason.

When asked separately whether their reasons for quitting a job were related to the coronavirus outbreak, 31% say they were. Those without a four-year college degree (34%) are more likely than those with a bachelor’s degree or more education (21%) to say the pandemic played a role in their decision.

For the most part, men and women offer similar reasons for having quit a job in the past year. But there are significant differences by educational attainment.

Among adults who quit a job in 2021, those without a four-year college degree are more likely than those with at least a bachelor’s degree to point to several reasons. These include not having enough flexibility to decide when they put in their hours (49% of non-college graduates vs. 34% of college graduates), having to work too few hours (35% vs. 17%) and their employer requiring a COVID-19 vaccine (21% vs. 8%).

There are also notable differences by race and ethnicity. Non-White adults who quit a job last year are more likely than their White counterparts to say the reasons include not having enough flexibility (52% vs. 38%), wanting to relocate to a different area (41% vs. 30%), working too few hours (37% vs. 24%) or their employer requiring that they have a COVID-19 vaccine (27% vs. 10%). The non-White category includes those who identify as Black, Asian, Hispanic, some other race or multiple races. These groups could not be analyzed separately due to sample size limitations.

Many of those who switched jobs see improvements

A majority of those who quit a job in 2021 and are not retired say they are now employed, either full-time (55%) or part-time (23%). Of those, 61% say it was at least somewhat easy for them to find their current job, with 33% saying it was very easy. One-in-five say it was very or somewhat difficult, and 19% say it was neither easy nor difficult.

For the most part, workers who quit a job last year and are now employed somewhere else see their current work situation as an improvement over their most recent job. At least half of these workers say that compared with their last job, they are now earning more money (56%), have more opportunities for advancement (53%), have an easier time balancing work and family responsibilities (53%) and have more flexibility to choose when they put in their work hours (50%).

Still, sizable shares say things are either worse or unchanged in these areas compared with their last job. Fewer than half of workers who quit a job last year (42%) say they now have better benefits, such as health insurance and paid time off, while a similar share (36%) says it’s about the same. About one-in-five (22%) now say their current benefits are worse than at their last job.

College graduates are more likely than those with less education to say that compared with their last job, they are now earning more (66% vs. 51%) and have more opportunities for advancement (63% vs. 49%). In turn, those with less education are more likely than college graduates to say they are earning less in their current job (27% vs. 16%) and that they have fewer opportunities for advancement (18% vs. 9%).

Employed men and women who quit a job in 2021 offer similar assessments of how their current job compares with their last one. One notable exception is when it comes to balancing work and family responsibilities: Six-in-ten men say their current job makes it easier for them to balance work and family – higher than the share of women who say the same (48%).

Some 53% of employed adults who quit a job in 2021 say they have changed their field of work or occupation at some point in the past year. Workers younger than age 30 and those without a postgraduate degree are especially likely to say they have made this type of change.

Younger adults and those with lower incomes were more likely to quit a job in 2021

Overall, about one-in-five non-retired U.S. adults (19%) – including similar shares of men (18%) and women (20%) – say they quit a job at some point in 2021, meaning they left by choice and not because they were fired, laid off or because a temporary job had ended.

Adults younger than 30 are far more likely than older adults to have voluntarily left their job last year: 37% of young adults say they did this, compared with 17% of those ages 30 to 49, 9% of those ages 50 to 64 and 5% of those ages 65 and older.

Experiences also vary by income, education, race and ethnicity. About a quarter of adults with lower incomes (24%) say they quit a job in 2021, compared with 18% of middle-income adults and 11% of those with upper incomes.

Across educational attainment, those with a postgraduate degree are the least likely to say they quit a job at some point in 2021: 13% say this, compared with 17% of those with a bachelor’s degree, 20% of those with some college and 22% of those with a high school diploma or less education.  

About a quarter of non-retired Hispanic and Asian adults (24% each) report quitting a job last year; 18% of Black adults and 17% of White adults say the same.

Note: Here are the questions used for this analysis, along with responses, and its methodology.

• Business & Workplace
• Coronavirus (COVID-19)
• COVID-19 & the Economy
• Income & Wages

Kim Parker is director of social trends research at Pew Research Center

Juliana Menasce Horowitz is an associate director of research at Pew Research Center

A look at small businesses in the U.S.

Majorities of adults see decline of union membership as bad for the u.s. and working people, a look at black-owned businesses in the u.s., from businesses and banks to colleges and churches: americans’ views of u.s. institutions, 2023 saw some of the biggest, hardest-fought labor disputes in recent decades, most popular.

1615 L St. NW, Suite 800 Washington, DC 20036 USA (+1) 202-419-4300 | Main (+1) 202-857-8562 | Fax (+1) 202-419-4372 |  Media Inquiries

Research Topics

• Age & Generations
• Economy & Work
• Family & Relationships
• Gender & LGBTQ
• Immigration & Migration
• International Affairs
• Internet & Technology
• Methodological Research
• News Habits & Media
• Non-U.S. Governments
• Other Topics
• Politics & Policy
• Race & Ethnicity
• Email Newsletters

ABOUT PEW RESEARCH CENTER  Pew Research Center is a nonpartisan fact tank that informs the public about the issues, attitudes and trends shaping the world. It conducts public opinion polling, demographic research, media content analysis and other empirical social science research. Pew Research Center does not take policy positions. It is a subsidiary of  The Pew Charitable Trusts .

Copyright 2024 Pew Research Center

Terms & Conditions

Privacy Policy

Cookie Settings

Reprints, Permissions & Use Policy