The purpose of this paper is to capture the investors' mood related to the COVID-19 pandemic and analyze its impact on the stock market returns.
To capture the investor mood related to the COVID-19 pandemic, the authors construct a unique COVID-19 fear index based on the Search Volume Index (SVI) from Google Trends (http://www.Google.com/trends/) of the search terms related to COVID-19 words and phrases as revealed by Google and Internet dictionaries. The COVID-19 fear index was used to investigate its impact on the stock market returns.
The study finds a strong negative association between COVID-19 fear and stock returns. Unlike other studies, the relationship is persistent for a significant period. This relationship is not found to reverse in the following days. The results also highlight that COVID-19 fear strongly impacts the stock market. The sentiment persists for a significant period and is not reversed soon, unlike the regular times in earlier studies.
The study is among the very few studies that constructed COVID-19 fear index using several Google search terms and captured its impact on the stock market returns.
Subramaniam, S. and Chakraborty, M. (2021), "COVID-19 fear index: does it matter for stock market returns?", Review of Behavioral Finance, Vol. 13 No. 1, pp. 40-50. https://doi.org/10.1108/RBF-08-2020-0215
Emerald Publishing Limited
Copyright © 2021, Emerald Publishing Limited
Moods and emotions felt today have an impact on people's choices and decisions with respect to the risk (Mann, 1992). People in a good mood are more optimistic in their choices and judgments than those in a bad mood. These affective states have an impact on asset pricing . Several researchers have documented various exogenous factors to capture the mood and are correlated with the stock returns. Saunders (1993) and Hirshleifer and Shumway (2003) examined using the weather, Kamstra et al. (2003) analyzed using the daylight, Yuan et al. (2006) investigated using the lunar phases, Edmans et al. (2007) studied using international soccer results, Drakos (2010) investigated the terrorism activity and Fernandez-Perez et al. (2020) used music to name a few. In this study, we examine how a retail investor's mood on the COVID-19 pandemic affects the stock market.
The current COVID-19 outbreak started in Wuhan in December 2019 and spread rapidly across the continents. World Health Organization declared the COVID-19 epidemic as a global pandemic on 11th March 2020. Since then, the transmission of corona infections has increased exponentially. As of 30th August 2020, the number of people infected was 5.97 million, with 183 thousand fatalities in the United States of America alone. The increasing number of infections led the government to take various precautionary measures to flatten the curve. The measures include sealing the borders, cancellation of international fights, ban of large gatherings, closure of educational institutions, entertainment avenues, gyms, restaurants, bars, etc. The outbreak of COVID-19 created a fear of health and survival issues caused by the virus. The fear of job losses and uncertainty during times of economic troubles has put the investors in a negative mental frame.
This study captures the sentiment induced by the COVID-19 pandemic and its impact on the stock returns. We construct a unique COVID-19 fear index based on the Search Volume Index (SVI) from Google Trends (http://www.google.com/trends/) of the search terms related to COVID-19 words and phrases as revealed by Google and Internet dictionaries. We construct the COVID-19 fear index from March 2020 to August 2020. To analyze the impact of COVID-19 fear on stock returns, the following stock market returns were considered: S&P 500 index, Russell 1000 ETF, Nasdaq 100 ETF and S&P 500 ETF. The study finds a strong negative association between COVID-19 fear and stock returns. Unlike other studies, the relationship is persistent for a significant period. This relationship does not reverse in the following days up to the next five days. Thus, in a nutshell, the results demonstrate the effect of sentiment during the global pandemic period and report that the impact of sentiment persists for a significant period and is not reversed soon, unlike the earlier studies during normal times.
Further, we extend the analysis to other COVID-19 affected countries such as India and Brazil to understand whether similar effects are present there too. The choice of countries is based on the rank of COVID-19 affected countries. The results of the study found a negative relationship between the COVID-19 fear index and their stock returns.
The study contributes to the literature in the following ways. First, it constructs a unique COVID-19 sentiment using the daily search items related to COVID-19. The idea of this paper is similar to the construction of the FEARS (Financial and Economic Attitude Revealed through Search) index proposed by Da et al. (2015), but this study focuses on the investor attention toward the COVID-19 pandemic. We construct the COVID-19 sentiment using the search words related to COVID-19 to capture investors' attention related to the virus. Da et al. (2015) construct the fears index based on the financial and economic terms.
Second, this study considers COVID-19 sentiment as a mood proxy, as it satisfies the criteria mentioned by Edmans et al. (2007). The COVID-19 sentiment drives the mood in a significant manner, and the effect is strong in the asset prices. It affects the mood of a large population, thereby influencing the investors, and it is correlated across the majority of individuals throughout the world.
Finally, the study extends the work of Salisu and Akanni (2020). They constructed a global fear index during the COVID-19 pandemic times using the number of reported cases and deaths. Our study extends their work by capturing the investor mood during the pandemic times using the Google search volume index.
The remainder of the paper is organized as follows. Section 2 reviews the literature pertaining to the COVID-19 pandemic. Section 3 describes the data and methodology used to construct the COVID-19 fear. Section 4 presents the empirical results and Section 5 concludes.
2. Literature review
The behavioral biases and their impact on asset prices have witnessed a burgeoning research. Several proxies have been documented to capture the mood or sentiment of investors. These are correlated with stock returns and as Rick and Lowenstein (2008) suggest, these emotions have an influence on risky decision making. Some studies also show that extreme and prolonged stress can adversely affect cognitive skills (Sapolsky, 1996). In an efficient market, such changes in mood or cognitive abilities are not supposed to have any additional influence on the stock returns after controlling for all the fundamental variables. Arbitrage operations will ensure that prices are affected only by the fundamentals. However, the alternate hypothesis of the influence of mood opens the door for behavioral finance. Due to limits to arbitrage, there can be the severe impact of behavioral biases on asset prices, and mispricing can be persistent as informed investors are not in a position to drive the prices toward fundamentals (Shleifer and Vishny (1997), Gromb and Vayanos (2002), Brunnermeier and Pedersen (2009)
In this study, we try to explore if there is any link between the COVID-19 sentiment and returns in the stock market. Following the literature on psychology, we assume that anxiety can give rise to a sense of uncertainty (Smith and Ellsworth, 1985; Ortony et al., 1988). The fear of job losses and uncertainty during times of economic troubles can put the investors in a negative mental frame. The study specifically tests whether psychological anxiety pertaining to COVID-19 has any relation to the stock market even after controlling for the fundamental factors that influence the economy and the market
The papers relating to the COVID-19 pandemic and its impact on the stock market are growing in the literature (Salisu and Vo, 2020; Topcu and Gulal, 2020; Akhtaruzzaman et al., 2021; to name a few). Narayan et al. (2021) examined the effect of government policies to counter the repercussion of COVID-19 and its impact on the stock market of G7 economies. They found that government policies were effective and the lockdown resulted in a cushioning effect. They also found that government policies have a positive impact on stock returns. Rizwan et al. (2020) estimated the systemic risk in eight of the COVID-19 affected countries. They found a sharp increase in the systemic risk during pandemic times. Zhang et al. (2020) found that the greater uncertainty of pandemic is associated with the higher volatility in the markets and are unpredictable.
The literature in capturing the COVID-19 fear and investor behavior are very few. Ortmann et al. (2020) used transactional level brokerage data of retail investors to analyze the investor behavior during COVID-19 times. Their study found that investors reduced the usage of leverage and increased their weekly trading intensity during COVID-19 times. Haroon and Rizvi (2020) investigated the sentiment estimated from the COVID-19 related news and analyzed its impact on stock markets' volatility. They found that panic generated news is associated with the increased volatility in the equity markets. Recently, Salisu and Akanni (2020) constructed a global fear index for the COVID-19. They used the reported cases and reported the death index to construct the fear index. They found that the fear index is a good predictor of stock return in OECD countries. Our study extends the work of Salisu and Akanni (2020) by measuring the fear index using the Google search volume index. The construction of the fear index using Google search volume index is helpful to measure the investor attention and the fear caused by COVID-19.
3. Data and methodology
The COVID-19 fear index is constructed using the search items from the Google trends described below in subsection 2.1. To capture the effect of fear caused by the corona pandemic on the stock market returns, the S&P 500 index, Russell 1000 ETF, Nasdaq 100 ETF and S&P 500 ETF is considered. The daily closing prices of the index were obtained from Thomson Reuters maintained by Refinitiv for the period March 2020–August 2020. The CBOE VIX measure, which is also known as investor fear gauge, is considered as an alternative to market sentiment measure (Baker and Wurgler, 2007), ADS  (Aruoba et al., 2009) business condition index is used to control for macroeconomic activities, economic policy uncertainty  measure to control the uncertainty related to policy measures and the change in the number of new COVID-19 cases were used as control variables.
Further, we extend our analysis to the other COVID-19 affected countries. For this purpose, we choose India and Brazil as they were second and third, respectively, in the list of worst affected countries (refer: https://www.worldometers.info/coronavirus/countries-where-coronavirus-has-spread/). The daily closing prices of stock indices of India (Nifty 500) and Brazil (Ibovespa) were obtained from Thomson Reuters for the same period. The daily frequency of CBOE VIX, Call money rate, Dividend yield, changes in FII and the change in the number of COVID-19 new cases were used as the control variables .
3.1 Construction of COVID-19 fear index
The COVID-19 fear index is constructed using the search terms related to the corona pandemic. We considered the search terms such as “COVID-19”, “Pandemic,” “CORONA Virus,” “Mask,” “Social Distancing,” “COVID-19 symptoms,” etc. Further, we also took the words commonly searched phrases and words displayed by the Google trends related to the COVID-19 pandemic. This resulted in 80 search items.
The search volume index of the 80 terms is obtained at the daily frequency from March 2020 to August 2020 from Google Trends . The daily log change in the search term is estimated as
The relationship between COVID-19 fear index and the stock market returns are estimated using the following model
4. Empirical findings
Table 1 reports the estimation results of the impact of COVID-19 fear on the aggregate market returns (S&P 500 returns). The coefficient of COVID-19 fear is found to be negative and significant after controlling for lagged returns and contemporaneous VIX, EPU, ADS and the change in the number of COVID-19 new cases. An increase in COVID-19 fear has a negative relationship with S&P returns. However, unlike Da et al., 2015, the day zero effect does not seem to reverse, and there exists a persistent negative impact in the following days. These results are revealing and are in line with the behavioral school, which suggests that over pessimism of investors can have a persistent impact and affect stock prices for significant periods of time. (Zouaoui et al., 2011).
We also examine the impact of COVID-19 sentiment on the highly liquid exchange-traded funds. For this purpose, we consider three high liquid ETFs, namely, Russell 1000 ETF, Nasdaq 100 ETF and S&P 500 ETF. The results reported in Table 2 are similar to the S&P 500 aggregate market index.
In an extension of the geographical context, we examine whether the COVID-19 fear index has an impact on the stock returns in two other countries, namely, India and Brazil, which were the second and third in the list of most affected countries. The results in panel A of Table 3 show that the COVID-19 fear sentiment has a negative impact on the Nifty 500 return. The coefficients of the fear on day t, t+1, and t+5 are significantly negative. The cumulative returns up to five days are significantly negative, suggesting the persistence of the negative effect.
Panel B of Table 3 shows that in Brazil's case, an increase in COVID-19 fear has a significantly negative impact on the IboVespa returns on day t+2 and cumulatively up to day t+5. The results of both India and Brazil are qualitatively similar to that of the US and provides evidence of the negative market reaction caused by the pandemic fear.
The study investigates the impact of COVID-19 fear on stock returns. We construct a unique COVID-19 fear index based on the Search Volume Index (SVI) from Google Trends (http://www.google.com/trends/) of the search terms related to COVID-19 words and phrases as revealed by Google and several Internet dictionaries. This COVID-19 fear index is used as a proxy to measure the retail investor's mood during this pandemic time. To analyze the impact of COVID-19 fear on stock returns, the following stock market returns were considered – S&P 500 index, Russell 1000 ETF, Nasdaq 100 ETF and S&P 500 ET. The results suggest a negative impact of COVID-19 fear on the stock returns, which seems to persist cumulatively even up to five days. Further, we investigate whether the fear of COVID-19 influences the other markets as well. For this, the analysis is done on two other countries, viz – India and Brazil, which ranked second and third in the list of most affected countries. The results are found to be similar to that of the US and provide evidence of a negative impact on the stock returns.
This relationship has a persistent effect on stock prices for a significant period of time. The findings may be useful for portfolio managers, as the COVID-19 fear index seems to be a good predictor of stock prices during pandemic times.
COVID-19 Fear and S&P 500 returns
|Ret(t)||Ret(t+1)||Ret(t+2)||Ret(t+1: t+2)||Ret(t+3)||Ret(t+4)||Ret(t+5)||Ret(t+1: t+5)|
|COVID-19 fear||−0.002*** (0.001)||−0.001 (0.000)||−0.003*** (0.000)||−0.004** (0.002)||−0.003*** (0.000)||−0.002* (0.000)||−0.003*** (0.000)||−0.010*** (0.003)|
|ADS||0.001 (0.012)||−0.018 (0.017)||0.011 (0.016)||−0.008 (0.017)||−0.004 (0.016)||0.009 (0.015)||0.004 (0.015)||0.002 (0.023)|
|EPU||−0.004 (0.006)||−0.020** (0.008)||0.017** (0.008)||−0.006 (0.010)||−0.012 (0.008)||0.013* (0.007)||−0.010 (0.007)||−0.016 (0.013)|
|VIX||−0.183*** (0.020)||0.004 (0.035)||−0.003 (0.030)||0.007 (0.065)||−0.075** (0.003)||−0.0306 (0.033)||−0.013 (0.032)||−0.105 (0.083)|
|Change in the number of COVID-19 new cases||−0.000 (0.000)||0.000 (0.000)||−0.000 (0.000)||−0.000 (0.000)||−0.000 (0.000)||−0.000 (0.000)||−0.000 (0.000)||−0.000 (0.000)|
|Returns (t)||−0.395*** (0.123)||0.136 (0.118)||−0.235 (0.294)||−0.328*** (0.122)||−0.352*** (0.115)||−0.002 (0.110)||−0.880** (0.373)|
|Returns(t−1)||−0.135 (0.066)||0.133 (0.091)||0.0432 (0.08)||0.175 (0.204)||−0.189** (0.090)||0.042** (0.086)||−0.237*** (0.082)||−0.205 (0.266)|
|Returns(t−2)||0.143 (0.064)||0.114 (0.087)||−0.199** (0.088)||−0.083 (0.200)||0.050 (0.086)||−0.1050 (0.082)||0.134* (0.077)||0.000 (0.255)|
|Returns(t-3)||−0.048 (0.062)||−0.089 (0.086)||−0.058 (0.081)||−0.135 (0.145)||−0.179** (0.084)||0.182** (0.079)||−0.070 (0.076)||−0.185 (0.223)|
|Returns(t−4)||−0.130** (0.062)||−0.021 (0.086)||−0.166** (0.083)||−0.181 (0.174)||0.217** (0.084)||−0.131 (0.082)||0.099 (0.076)||0.0144 (0.211)|
|Returns(t−5)||0.090 (0.059)||−0.212** (0.081)||0.343*** (0.077)||0.125 (0.201)||−0.1321* (0.084)||0.208*** (0.076)||−0.0810 (0.071)||0.115 (0.224)|
|Constant||−0.009 (0.004)||0.005 (0.005)||−0.003 (0.005)||0.003 (0.006)||−0.002 (0.005)||−0.0042 (0.005)||−0.005 (0.005)||−0.007 (0.009)|
Note(s): This table presents the regression results of S&P 500 returns to COVID-19 fear. The dependent variables are contemporaneous returns (Column 1), and future S&P returns up to five days (Columns (2), (3), (5), (6), (7)) and cumulative returns up to 2days (column (4)) and 5days (column (8)). The control variables include lagged returns and changes in ADS, EPU, VIX, change in the COVID-19 new cases and weekday dummies. The bootstrapped standard errors are reported in parenthesis. ***, ** and * denotes 1, 5 and 10% significance level
COVID-19 fear and returns to the other asset classes
|Ret(t)||Ret(t+1)||Ret(t+2)||Ret(t+5)||Ret (t+1 : t+5)|
|Panel A: COVID-19 fear and Russell 1000 ETF returns|
|COVID-19 fear||−0.0023*** (0.000)||−0.001 (0.000)||−0.0032*** (0.001)||−0.003*** (0.001)||−0.007*** (0.002)|
|Panel B: COVID-19 fear and NASDAQ ETF returns|
|COVID-19 fear||−0.002*** (0.001)||−0.001 (0.001)||−0.003** (0.001)||−0.002** (0.001)||−0.007*** (0.001)|
|Panel C: COVID-19 fear and S&P 500 ETF|
|COVID-19 Fear||−0.002*** (0.001)||−0.001 (0.001)||−0.003*** (0.001)||−0.003*** (0.001)||−0.008*** (0.002)|
Note(s): This table presents the regression results of Russell 1000 ETF returns (Panel A), Nasdaq 100 ETF returns (Panel B) and S&P 500 ETF (Panel C) to COVID-19 fear. The dependent variables are contemporaneous returns (Column 1), and future S&P returns up to five days (Columns (2), (3) and (5)) and cumulative returns up to 5 days (column (5)). The control variables include lagged returns and changes in ADS, EPU and VIX. The bootstrapped standard errors are reported in parenthesis. ***, ** and * denotes 1, 5 and 10% significance level
COVID-19 fear and stock returns of other economies
|Ret(t)||Ret(t+1)||Ret(t+2)||Ret(t+5)||Ret (t+1: t+5)|
|Panel A: COVID-19 fear and India nifty 500 returns|
|COVID-19 fear||−0.00056** (0.000)||−0.00128 (0.000)||−0.00203** (0.000)||−0.003*** (0.000)||−0.009*** (0.001)|
|Panel B: COVID-19 fear and Brazil Ibovespa index|
|COVID-19 Fear||0.001 (0.112)||−0.0006 (0.002)||−0.0037*** (0.001)||−0.00142 (0.001)||−0.0069*** (0.002)|
Note(s): This table presents the regression results of India's nifty 500 Returns (Panel A) and Brazil Ibovespa Index returns (Panel B) to COVID-19 fear. The dependent variables are contemporaneous returns (Column 1), and the future index returns up to five days (Columns (2), (3) and (5)) and cumulative returns up to 5days (column (5)). The daily frequency of CBOE VIX, Call money rate, Dividend yield, changes in FII and the change in the number of COVID-19 new cases were used as the control variables. The bootstrapped standard errors are reported in parenthesis. ***, ** and * denotes 1, 5 and 10% significance level
List of search words and trending Coronavirus questions (provided by Google trends)
|COVID||Work from home|
|Quarantine||Acute respiratory distress syndrome|
|Social distancing||COVID breakout|
|Fomite||Shortness of breath|
|Community spread||Loss of taste|
|Contact tracing||Loss of smell|
|Mortality rate||Hand wash|
|Flatten the curve||Can you get corona more than once|
|Respirator||Is corona virus getting better|
|Ventilator||What are the symptoms of corona virus|
|Flu||How is corona virus transmitted|
|Spanish flu||COVID death|
|Sars||What percentage of people die from corona virus|
|Asymptomatic||How long after corona virus are you contagious|
|Vaccine||How long does it take to get results from the corona virus test|
|Clinical trial||Can you get corona virus more than once|
|Containment area||What are the symptoms of corona virus|
|Hydroxycholoroquine||Corona virus airborne|
|Incubation period||Early signs of corona virus|
|Novel coronavirus||Economic chaos|
|Physical distancing||Economic uncertainty|
|Person to person transmission||Plasma therapy|
|Ppe kit||Respiratory droplets|
For extensive literature, refer to Hirshleifer, 2001.
The data is available at https://www.philadelphiafed.org/research-and-data/real-time-center/business-conditions-index
The data is available at https://www.policyuncertainty.com/
As the ADS business condition index and Economic policy uncertainty data was not available for India and Brazil at a daily frequency
The search queries on most of these terms were negligible before March 2020. After the first reported death in the US on 29th February 2020, the COVID-19-related search queries gained momentum.
Aruoba, S.B., Diebold, F.X. and Scotti, C. (2009), “Real-time measurement of business conditions”, Journal of Business and Economic Statistics, Vol. 27, pp. 417-27.
Akhtaruzzaman, M., Boubaker, S. and Sensoy, A. (2021), “Financial contagion during COVID–19 crisis”, Finance Research Letters, Vol. 38, p. 101604.
Baker, M. and Wurgler, J. (2007), “Investor sentiment in the stock market”, Journal of Economic Perspectives, Vol. 21, pp. 129-152.
Brunnermeier, M. and Pedersen, L. (2009), “Market liquidity and funding liquidity”, Review of Financial Studies, Vol. 22, pp. 2201-2238.
Da, Z., Engelberg, J. and Gao, P. (2015), “The sum of all FEARS investor sentiment and asset prices”, The Review of Financial Studies., Vol. 28, pp. 1-32.
Drakos, K. (2010), “Terrorism activity, investor sentiment, and stock returns”, Review of Financial Economic, Vol. 19, pp. 128-135.
Edmans, A., García, D. and Norli, Y. (2007), “Sports sentiment and stock returns”, The Journal of Finance, Vol. 62, pp. 1967-1998.
Fernandez-Perez, A., Garel, A. and Indriawan, I. (2020), “Music sentiment and stock returns”, Economics Letters, Vol. 192, C, p. 109260.
Gromb, D. and Vayanos, D. (2002), “Equilibrium and welfare in markets with financially constrained arbitrageurs”, Journal of Financial Economics, Vol. 66, pp. 361-407.
Haroon, O. and Rizvi, S.A.R. (2020), “COVID-19: media coverage and financial markets behavior—a sectoral inquiry”, Journal of Behavioral and Experimental Finance, p. 100343.
Hirshleifer, D. (2001), “Investor psychology and asset pricing”, The Journal of Finance, Vol. 56, pp. 1533-1597.
Hirshleifer, D. and Shumway, T. (2003), “Good day sunshine: stock returns and the weather”, The Journal of Finance, Vol. 58, pp. 1009-1032.
Kamstra, M.J., Kramer, L.A. and Levi, M.D. (2003), “Winter blues: a sad stock market cycle”, American Economic Review, Vol. 93, pp. 324-43.
Mann, L. (1992), “Stress, affect, and risk taking”, in Yates, J.F. (Ed.), Wiley Series in Human Performance and Cognition. Risk-Taking Behavior, John Wiley & Sons, pp. 202-230.
Narayan, P.K., Phan, D.H.B. and Liu, G. (2021), “COVID-19 lockdowns, stimulus packages, travel bans, and stock returns”, Finance Research Letters, Vol. 38, p. 101732.
Ortmann, R., Pelster, M. and Wengerek, S.T., (2020), “COVID-19 and investor behavior”, available at SSRN 3589443.
Ortony, Clore and Collins (1988), “The cognitive structure of emotion” January, Contemporary Sociology, Vol. 18 No. 6.
Rick, S. and Loewenstein, G. (2008), “The role of emotion in economic behavior, from handbook of emotions”, in Lewis, M., Haviland-Jones, J.M. and Barrett, L.F. (Eds), 3rd ed., The Guilford Press, pp. 138-156.
Rizwan, M.S., Ahmad, G. and Ashraf, D. (2020), “Systemic risk: the impact of COVID-19”, Finance Research Letters, Vol. 36, p. 101682.
Salisu, A.A. and Vo, X.V. (2020), “Predicting stock returns in the presence of COVID-19 pandemic: the role of health news”, International Review of Financial Analysis, Vol. 71, p. 101546.
Salisu, A.A. and Akanni, L.O. (2020), “Constructing a global fear index for the COVID-19 pandemic”, Emerging Markets Finance and Trade, Vol. 56 No. 10, pp. 2310-2331.
Sapolsky, R.M. (1996), “Why stress is bad for your brain”, Science, Vol. 273, pp. 749-750.
Saunders, E.M. Jr (1993), “Stock prices and wall street weather”, American Economic Review, Vol. 83, pp. 1337-1345.
Shleifer, A. and Vishny, R. (1997), “The Limits of arbitrage”, Journal of Finance, Vol. 52, pp. 35-55.
Smith and Ellsworth (1985), “Patterns of cognitive appraisal in emotion”, Journal of Personality and Social Psychology, Vol. 48 No. 4, pp. 813-38.
Topcu, M. and Gulal, O.S. (2020), “The impact of COVID-19 on emerging stock markets”, Finance Research Letters, Vol. 36, p. 101691.
Yuan, K., Zheng, L. and Zhu, Q. (2006), “Are investors moonstruck? Lunar phases and stock returns”, Journal of Empirical Finance, Vol. 13, pp. 1-23.
Zhang, D., Hu, M. and Ji, Q. (2020), “Financial markets under the global pandemic of COVID-19”, Finance Research Letters, Vol. 36, p. 101528.
Zouaoui, M., Nouyrigat, G. and Beer, F. (2011), “How does investor sentiment affect stock market crises? Evidence from panel data”, Financial Review., Vol. 46, pp. 723-747.
Baker, M. and Wurgler, J. (2006), “Investor sentiment and the cross‐section of stock returns”, The Journal of Finance, Vol. 61, pp. 1645-1680.
The authors are thankful to the Editor and the anonymous reviewers for their valuable comments and suggestions for improvement.