How does investor sentiment affect stock market crash risk? Evidence from Asia-Pacific markets

Nguyen, An Tuan; Nguyen, Nhung Thi Article How does investor sentiment affect stock market crash risk? Evidence from Asia-Pacific markets Cogent Economics & Finance Provided in Cooperation with: Taylor & Francis Group Suggested Citation: Nguyen, An Tuan; Nguyen, Nhung Thi (2024) : How does investor sentiment affect stock market crash risk? Evidence from Asia-Pacific markets, Cogent Economics & Finance, ISSN 2332-2039, Taylor & Francis, Abingdon, Vol. 12, Iss. 1, pp. 1-17, https://doi.org/10.1080/23322039.2024.2422959 This Version is available at: https://hdl.handle.net/10419/321653 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Cogent Economics & Finance ISSN: 2332-2039 (Online) Journal homepage: www.tandfonline.com/journals/oaef20 How does investor sentiment affect stock market crash risk? Evidence from Asia-Pacific markets An Tuan Nguyen & Nhung Thi Nguyen To cite this article: An Tuan Nguyen & Nhung Thi Nguyen (2024) How does investor sentiment affect stock market crash risk? Evidence from Asia-Pacific markets, Cogent Economics & Finance, 12:1, 2422959, DOI: 10.1080/23322039.2024.2422959 To link to this article: https://doi.org/10.1080/23322039.2024.2422959 © 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group View supplementary material Published online: 04 Nov 2024. Submit your article to this journal Article views: 1840 View related articles View Crossmark data Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=oaef20 FINANCIAL ECONOMICS | RESEARCH ARTICLE How does investor sentiment affect stock market crash risk? Evidence from Asia-Pacific markets An Tuan Nguyen and Nhung Thi Nguyen Faculty of Finance and Banking, VNU University of Economics and Business, Hanoi, Vietnam ABSTRACT This study aims to examine the effect of investor sentiment on stock market crash risk in the Asia–Pacific region. The research employs principal components analysis (PCA) to construct an investor sentiment index, while the Method of Moments Quantile Regression (MMQR) is used to analyze monthly data of 16 Asia-Pacific stock markets. The findings show that investor sentiment positively impacts on crash risk in the middle to higher quantities. Moreover, regional sentiment significantly increases stock market crash risk, particularly at higher quantiles, while local sentiment generally reduces crash risk at the lower to middle quantiles. Besides, the magnitude and direction impact of investor sentiment on stock market crash risk is heterogeneous across market levels. Specifically, the results indicate that at higher quantiles of risk, investor sentiment increases crash risk in developed and emerging markets, while it decreases crash risk in frontier markets. IMPACT STATEMENT This paper not only provides support for behavioral theories but also have implications for global investors, portfolio managers, and policymakers. ARTICLE HISTORY Received 9 July 2024 Revised 13 October 2024 Accepted 24 October 2024 KEYWORDS Stock market crash risk; market risk; investor sentiment; Asia-Pacific equity markets; crash risk SUBJECTS Psychological Science; Finance; Economic Psychology JEL CLASSIFICATION CODES D53; G21; G40 1. Introduction Investing in the stock market carries certain risks due to specific risks or market risks, with the possibility of occurrence of such risks being higher during a financial crisis (Wang et al., 2009). Financial crises have historically had far-reaching economic, social, and political ramifications (Fu et al., 2020). Stock market crashes, which are rapid and often unforeseen declines in stock prices, can be a side effect of a major catastrophic event, economic crisis, or the collapse of a long-term speculative bubble (Kustina et al., 2024). In particular, stock and bond markets play an essential roles in the financial stability of national capital markets (Zhou et al., 2022). Past decades have experienced many stock market crashes, such as the collapse in 1929 in the United States, the 2018 financial crisis leading to a drop in the value of subprime mortgage stocks, and the COVID-19 pandemic in March 2020. Understanding the risk of a stock price collapse is critical for portfolio investment, risk management, and stakeholder safety monitoring (Vo, 2020). In theory, a stock market crash is defined as a sudden decline in stock returns in most stocks, leading to a significant decrease in investors’asset values (Wang et al., 2009). Mishkin and White (2002) state that a stock market crash quantifies a 20% decline in the market index over a period of time (between a day and a year). Patel and Sarkar (1998) define ‘collapse’as a significant drop in a market index relative to its historical maximum over an observed period, such as one or two years. These authors classify a stock market crash as a relative decline in the regional price index of more than 20% for developed markets and more than 35% for emerging markets. Furthermore, Hong and Stein (2003) argue that a ‘crash’ CONTACT Nhung Thi Nguyen [email protected] Faculty of Finance and Banking, VNU University of Economics and Business, Hanoi, Vietnam Supplemental data for this article is available online at https://doi.org/10.1080/23322039.2024.2422959 ß2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent. COGENT ECONOMICS & FINANCE 2024, VOL. 12, NO. 1, 2422959 https://doi.org/10.1080/23322039.2024.2422959 must consist of three elements, including: (i) a collapse, characterized by an abnormally large movement in the market without a corresponding major mass news event; (ii) that this large price change is negative; and (iii) the crash constitutes a market-wide ‘contagion’phenomenon. This means that a crash is not only a sudden drop in the price of a stock but also a fall strongly correlated with the price of an entire class of stocks. In addition, crash risk is defined as the conditional deviation of the return distribution, which shows an asymmetry in risk and is significant for investment and risk management choices (Kim et al., 2014). According to Zhang et al. (2021), crash risk is a type of extreme risk or tail risk, which can be also defined as extreme events, whose probability of occurring is shown in the tail of a distribution and is capable of influencing valuation results. In general, stock market crash risk is an extreme type of risk that indicates the likelihood of a sudden market crash caused by cumulative negative information that has been made public for an extended period. A stock market crash risk comes from the stock price crash risk of firms where managers tend to hide accumulated bad news over a long period (Habib et al., 2018). If company executives successfully prevent the flow of negative information into the market, the distribution of stock returns will be disproportionate (Hutton et al., 2009; Kothari et al., 2009). When the amount of bad information crosses a threshold, it is revealed to the market immediately, leading to a stock price crash. In addition, stock price crash risk is also affected by individualism, especially during the global financial crisis (An et al., 2018). This impact can be eliminated through enhanced financial information transparency. The Efficient Market Hypothesis (EMH) of Fama (1970) states that stock prices react to information and interpret relevant information to form good and bad expectations about stock price movements. As a result, the buy and sell orders of all participants in the market are pooled to establish the equilibrium price of shares at a certain point in time. However, investors are always constrained in their reasoning and cannot effectively evaluate all types of information. This means that their biases can lead to an increase in market sentiment. Stock markets are prone to price bubbles, leading to a crash in the market. Moreover, the asymmetric information theory explains that information asymmetry occurs when different people know different things. This means that an asymmetric capital market emerges when one market player (management) knows more about the company’s internal information while investors or shareholders know little about the business’s information, leading to an incentive for executives to take action. Cao et al. (2002) propose the ‘information blockade’model to account for price collapse through information congestion and the asymmetric release of information in the stock market at the cost of setting up fixed trades. The increasing price trend drives well-informed investors to initiate aggressive trading in line with this pattern. Conversely, less knowledgeable investors often doubt the true nature and accuracy of the signal, thus delaying trade until the price falls. As a result, price corrections are unavoidable if the economic outlook turns negative and investors join the market with less knowledge. Furthermore, because investors might postpone trading until price movements corroborate their private signals, insider trading generates even more market news. In particular, inheriting the signal theory of French et al. (1987) and Ross (1973), Campbell and Hentschel (1992) propose the volatility feedback effect to explain stock crash price risk. In particular, a large volume of good news indicates an increase in market volatility. Therefore, the immediate benefit of positive news is somewhat offset by a rise in the risk premium. On the other hand, when large amounts of bad news emerge, the direct impact and risk-offset effect will go in the same direction, leading to an amplification of the influence. Extreme price movements might induce investors to rethink market volatility and raise the necessary risk compensation, which lowers the equilibrium price by building up the effect of adverse news and balancing the influence of good news, resulting in a negative skewness about return (Hutton et al., 2009). Besides the above-mentioned arguments of traditional finance theories, behavioral finance theory states that a stock market crash risk also results from psychological factors (Zouaoui et al., 2011). According to Blajer-GołeRbiewska et al. (2018) and Lucey and Dowling (2005), asset prices are not only driven by reasonable expected returns but also irrational decisions that are affected by investor emotions, sentiments, or states of mind. In fact, investor sentiment is defined as participants’expectations regarding future cash flows (returns) and investment risk (De Long et al., 1990). When sentiment is considered as high, stocks have comparatively poor subsequent returns, and when sentiment is low, these cross-sectional patterns are inverted (Baker & Wurgler, 2006). Furthermore, sentiment-based mispricing is caused by uninformed demand from investors, noise traders, and an arbitrage limit. Therefore, there 2 A. T. NGUYEN AND N. T. NGUYEN have been several empirical investigations on the role of investor sentiment in stock market crash risk which can be divided into three main topics as follows: (i) Studies on the effect of market sentiment on individual stock price crash risk, such as those of Cui and Zhang (2020), Yin and Tian (2017), and Wu et al. (2021); (ii) Studies on the effect of firm-specific investor sentiment on stock price crash, such as those of Alnafea and Chebbi (2022), Fan et al. (2021), Fu et al. (2021), and Shin and Choi (2022); and (iii) Studies on investor sentiment in early warning systems of stock market crashes, such as those of Fu et al. (2020) and Zouaoui et al. (2011). However, it can be seen that most previous studies focus on the impact of investor sentiment on stock price crash risk at the firm level but ignore the country level. One of the few closest studies focused on cross-sectional countries is that of Kustina et al. (2024) but investor sentiment indexes are measured by a simple formula based on the highest high, lowest low, and closing price of an asset (Zhou, 2018), which does not fully reflect market sentiment as the composite index of Baker and Wurgler (2006). Besides, these authors only use logistic regression and OLS regression, which leads to bias in results because panel data requires consideration of cross-sectional dependence and stationarity. Furthermore, previous studies have not considered market development as one of the factors influencing the relationship between investor sentiment and stock market crash risk. Meanwhile, Chui et al. (2010) claim that market development and market integrity can affect information flow and market efficiency. Motivated by the above-mentioned research gap, this research aims to investigate the impact of investor sentiment on stock market crash risk among Asia–Pacific markets. In fact, Asia-Pacific markets are sub-categorised into frontier, emerging, and developed markets, which allows the research to compare this effect among markets of different development levels. Despite their ongoing evolution and improvements, Asia-Pacific stock markets face challenges, such as undeveloped market structures, a low number of institutional investors, and incomplete transparency, highlighting the need for research in this region. In addition, this study also employs the Method of Moments Quantile Regression (MMQR) on panel data of 16 stock markets spanning from January 2006 to December 2023 to examine the dependence of the relationship between investor sentiment and the probability of stock market crash by levels of market development. This research makes the following two contributions. Firstly, this study adds to the literature review by showing that investor sentiment impacts the different quantiles of crash risk in Asia-Pacific stock markets. Moreover, the magnitude and direction impact of investor sentiment on stock market crash risk is heterogeneous across market levels. Secondly, this study proposes some recommendations for policymakers and investors for a stable market and an effective portfolio diversification strategy, respectively. The remainder of this article is structured as follows: Section 2 provides a literature review on the effect of investor sentiment on the crash risk of the stock market. Section 3 explains methods of measuring variables and collecting and analyzing the data. Section 4 describes empirical results before they are discussed in section 5. Finally, conclusions with implications, limitations, and orientation for future studies are presented in section 6. 2. Literature reviews Topics about the impact of investor sentiment on stock market crash risk have attracted the attention of many scholars. There are three strands of literature review, including (i) the effect of market sentiment on individual stock price crash risk; (ii) the effect of firm-specific investor sentiment on stock price crash risk; and (iii) the effect of investor sentiment in early warning systems of stock market crashes. In terms of the effect of market sentiment on individual stock price crash risk, Yin and Tian (2017), based on the method proposed by Baker and Wurgler (2006), construct a market sentiment index and investigate the effect of market sentiment on stock price crash risk in China. The results demonstrated that there is evidence of positive impacts of investor sentiment on the future stock price crash risk, and this nexus is strengthened by poorer financial reports, the absence of short-sales constraints, and the bull market state. Similar conclusions are reached by Cui and Zhang (2020) for the U.S. enterprises between 1991 and 2014. Besides, Wu et al. (2021) also demonstrate rising holistic investor sentiment in the current period can increase stock price crash risk in the next period in both Shanghai and Shenzhen A-share markets. Moreover, their results reveal that investor pessimism will increase stock price crash risk COGENT ECONOMICS & FINANCE 3 in the Shenzhen A-share market from the perspective of heterogeneous sentiment. With a large sample of A-share listed companies on the Shanghai and Shenzhen Stock Exchanges in the period of 2004– 2019, Bashir et al. (2024) outline the mediating role of analyst herding in deepening market sentiment which results in crash risk. However, the market-wide investor sentiment index is invariant in the crosssection and cannot correctly represent investor sentiment with individual firms, making it inappropriate for dealing with firm-level issues (Aboody et al., 2018). Regarding the impact of firm-specific investor sentiment on stock price crash risk, this effect is developed in an attempt to address the limitations of the market-wide sentiment index. Fu et al. (2021) find that firm-specific investor sentiment is significantly and positively related to the possibility of crash occurrence in stock prices. Additionally, the effect of firm-specific investor sentiment on crash risk is more pronounced for firms with worse liquidity, which is supported by Alnafea and Chebbi (2022) who argue that firms with worse liquidity have a considerably greater positive influence of investor sentiment on future crashes than firms with better liquidity. Shin and Choi (2022) indicate that firms listed on the KSE from 2011 to 2019 that also have high levels of foreign ownership reduce the high stock price crash risk attributable to high sentiment. Besides the above, the impact of investor sentiment on future stock price crash risk is more significant for stocks eligible for margin trading (Fan et al., 2021). In addition, Zhang et al. (2023) show that stock price synchronicity which is defined as firm-specific information or noise in the stock price, can harm investor attitude to obtain firm-specific information, leading to a higher crash risk. Concerning the impact of investor sentiment on stock market crash risk, Zouaoui et al. (2011) use a logit model to show investor sentiment has a significantly positive influence on the probability of occurrence of stock market crash. Additionally, they reveal that investor sentiment has a stronger impact on stock markets in countries where herding and overreaction are common, while institutional involvement is limited. This finding is entirely supported by Zhang et al. (2019) who confirm that investor sentiment is a better predictor of the stock market crisis within a 1-year horizon than macroeconomic variables, and Fu et al. (2020) who focus on the important role of sentiment factors in early warning models. Recently, Kustina et al. (2024), who define country index crash risk as the possibility of the significant and rapid decline of the stock index, give evidence of a negative impact of investor sentiment and exchange rate on the country index crash risk. Furthermore, a higher net foreign trading value does not improve the impact of investor sentiment but reduces the influence of exchange rate volatility on the country index crash risk. 3. Methodology 3.1. Measuring variables 3.1.1. Dependent variable As regards methods used to measure stock market crash risk, there are two approaches, including (i) CMAX index; and (ii) Asymmetric returns. The CMAX index measures current prices against the highest in the past 12 or 24 months, with a crash occurring when it falls below two standard deviations from its historical average (Patel & Sarkar, 1998). To assess return asymmetry, Chen et al. (2001) use the negative coefficient of skewness (NCSKEW) and down-to-up volatility (DUVOL), while Liu et al. (2021) enhance the conditional skewness with the GARCH-S model for daily deviations over six months. In this study, asymmetric return measures, specifically NCSKEW and DUVOL, are chosen over CMAX for detecting stock market crash risk due to their focus on left-tail risk and volatility asymmetry. NCSKEW captures extreme negative returns, making it more effective at identifying early warning signs of market downturns, while DUVOL emphasizes market reactions to negative news by comparing the volatility on down days vs. up days. Both measures rely on daily data, allowing for quicker responses to emerging risks compared to CMAX, which focuses on longer-term trends and may overlook short-term fluctuations. This sensitivity to market dynamics, particularly in volatile conditions, makes NCSKEW and DUVOL more suitable for assessing crash risk. Therefore, the negative coefficient of skewness (NCSKEW) and down-to-up volatility (DUVOL) are used to measure stock market crash risk. The idiosyncratic daily market return is defined as 4 A. T. NGUYEN AND N. T. NGUYEN Wi,t¼ln 1 þRi,t ðÞ ,whereR i,tis the return of stock market index i on day t. The negative coefficient of skewness (NCSKEW) is the ratio of the third moment of stock market returns over the standard deviation of stock market returns raised to the third power and then multiplied by −1, as shown below: NCSKEWi,T¼− nðn−1Þ3=2PW3 i,t n−1 ðÞ ðn−2ÞPW2 i,t  3=2 where n is the trading days of stock market index i in month T. Adding a negative sign on the right side of the equation will make the negative coefficient of skewness positively correlate with the stock price crash. In other words, the higher the negative skewness coefficient is, the higher the stock price crash risk is. The down-to-up volatility (DUVOL) is calculated as the following formula: DUVOLi,T¼ln ðnup −1ÞPdownW2 i,t ðndown −1ÞPupW2 i,t where nup and ndown are the up and down days. For returns in month T, we separate all the days with returns below the monthly mean (down days) from those with returns above the monthly mean (up days) and calculate the standard deviation for each of these sub-samples separately. Then, the DUVOL measure is the log of the ratio of the standard deviation of the down days to the standard deviation of the up days. An increase in DUVOL corresponds to a stock being more likely to crash and vice versa. 3.1.2. Independent variable From the literature review, there are three methods to build an investor sentiment index, including: (i) The survey-based approach allows for the analysis of different investor groups but is limited by time and space; (ii) The text-based method provides high-frequency indicators but risks inaccurate correlations due to vast data and cross-country complexities due to language differences; (iii) The market-based method, favored for its data availability, struggles to separate rational from psychological behavior. Since a composite sentiment index better captures investor irrationality compared to a single sentiment index (Yang & Gao, 2014), we apply the Baker and Wurgler (2006) method. This approach involves capturing the first principal component of sentiment proxies to measure investor sentiment across Asia- Pacific stock markets. The process of calculating the investor sentiment index is shown in the following steps: (i) Step 1: Collect stock market-level data on technical indicators. This study utilizes four key technical indicators that can represent investor sentiment on Asia-Pacific stock markets, including: (i) The relative strength index (RSI) is a technical indicator that helps investors detect overbought or oversold conditions in the market, to show the investors’beliefs in the markets; (ii) Williams %R (WRI) measures market extremes by comparing the current closing price to the highest and lowest prices over a set period, (iii) The psychological line index (PLI) measures investor volatility during market fluctuations and captures short-term price reversals, reflecting short-term trends and psychological stability; and (iv) The share turnover velocity (VOL) can capture the liquidity of a stock market and investors’heterogeneous beliefs on a stock market, as higher turnover suggests optimism and expectations of rising prices. These four indicators are chosen because of their high pairwise correlations (Supplementary Appendix 1). Table 1 presents a detailed description of sentimental proxies. (ii) Step 2: Use the principal components analysis (PCA) approach to assign weights to all four proxies of investor sentiment in Table 1 and then construct a composite investor sentiment index based on the first principal component (Fi) for each stock market i. investor sentiment indicator (SENTi,t) is calculated as: SENTi,t¼Fi,1RSIi,tþFi,2WRIi,tþFi,3PLIi,tþFi,4VOLi,t COGENT ECONOMICS & FINANCE 5 Besides, sentiment contagion spreads across markets within a geographic region, forming regional sentiment (Baker et al., 2012). Regional sentiment is derived by performing the PCA on investor sentiment indicators from 16 countries in the sample, while local sentiment for each market is calculated as the residual after orthogonalizing total sentiment concerning regional sentiment. Detailed results of the PCA approach are shown in Supplementary Appendix 2. 3.1.3. Control variables This study employs an expanded set of market characteristics and macroeconomic conditions as control variables to isolate the ‘irrational’component of market sentiment measures, acknowledging that changes in these factors can affect overall economic conditions and market sentiment. To be precise, the four factors considered include market return (RET), market volatility (SIGMA), inflation (INF), and interest rate (INT). Market return is determined by the change in the natural logarithm of the stock index from month t to month t-1, while market volatility is assessed through the standard deviation of daily returns in month t. Inflation is represented by the change in the natural logarithm of the monthly consumer price index, and the monthly money market rate indicates the interest rate (INT). The main variables are summarized in Supplementary Appendix 3. 3.2. Data and sampling This study uses panel data of 16 stock markets (Supplementary Appendix 4) with 216 months from January 2006 to December 2023, giving a sample of data that includes 16 216 ¼3456 observations. Since the stock market index changes daily, but macroeconomic data is recorded monthly, quarterly, or yearly, it is appropriate to determine monthly data. Secondary data on the stock index extracted from Bloomberg consists of 16 stock market indices classified from the MSCI catalog and internationally recognized as benchmark indices, which are the most closely watched by analysts in the Asia-Pacific region. In addition, macro factors are compiled from the International Financial Statistics of IMF and Investing.com. Table 2 shows the summary statistics for the entire sample, such as minimum values, standard deviation, mean value, maximum values, and number of observations used in the study. The main variable of interest is stock market crash risk which is measured by the negative coefficient of skewness (NCSKEW) and down-to-up volatility (DUVOL). NCSKEW has an average of −0.020, a slight negative skewness in stock returns, suggesting that extreme negative returns are more likely than extreme positive returns. DUVOL also reflects downside risk, with a mean of −0.066, implying that downside volatility tends to be more pronounced than upside volatility, making losses sharper than gains on average. Table 1. Description of sentiment proxies. Proxy Code Measure References Relative strength index RSI RSIi,t¼RSi,t 1þRSi,t100 where RSIi,t¼P14 t¼1max 0,Pi,t−Pi,t−1 ðÞ P14 t¼1max 0,Pi,t−1−Pi,t ðÞ , and Pi,tis closing price of stock market at month t. Zhou et al., 2023 %William R indicator WRI WRIi,t¼Phighest,i,t−Pi,t Phighest,i,t−Plowest,i,t−100 where Phighest,i,tand Plowest,i,tare the highest high and lowest low prices of stock index i, the past n ¼14 periods are used, and closing price is the closing price today. Kustina et al., 2024; Zhou, 2018 Psychological line index PLI PLIi,t¼100 P 12 t¼1 Max 0,Pi,t−Pi,t−1 ðÞ Pi,t−Pi,t−1 no =12 where Pi,tis closing price of stock market i at month t. Yang & Gao, 2014; Zhou et al., 2023 Share turnover velocity VOL VOLi,t¼1 12 P 11 j¼0 TURNi,t-j Where TURNi,t¼Total of shares tradei,t Maket capitalizationi,t, and TURNi,trepresents share turnover velocity of equity market i at month t. Baker & Wurgler, 2006 Source: Authors. 6 A. T. NGUYEN AND N. T. NGUYEN The mean value of investor sentiment (SENT) is 0.000, implying that the data contains equal occurrences of positive and negative sentiment. The large standard deviation of 1.607 and the range from −4.693 to 4.070 indicate significant swings in market sentiment, reflecting periods of both extreme pessimism and optimism. Countries in the sample have average levels of inflation (INF) of 0.004, reflecting a low inflationary environment, while interest rates (INT) average 4.1%, but exhibit significant variation, ranging from −0.001 to 0.800, reflecting different economic conditions. On average, market returns are slightly positive at 0.5%, but with noticeable fluctuations, as indicated by a standard deviation of 0.059. Finally, market volatility (SIGMA) averages 1.0%, with minimum and maximum values of −0.027 and 0.07, respectively, suggesting that while most markets are relatively stable, some experience considerable fluctuations. 3.3. Econometrics approach For the empirical modeling, we employ panel co-integration techniques which are based on (i) checking the cross-sectional dependence (CD); (ii) testing the order of integration corresponding to each variable; and (iii) the existence of long-run co-integration among the variables is checked using panel co-integra- tion tests. Once the long-run cointegration is confirmed, we need to estimate the long-run coefficients by innovative estimate methodology as MMQR, to investigate the relationship between variables. Machado and Silva (2019) came up with this approach. The reason for using the MMQR methodology is that it outperforms other methods, such as conventional panel quantile regression, DOLS, and FMOLS. MMQR is suited since other linear estimating approaches cannot handle distributions of data and hence only address averages. Moreover, basic quantile regression is inadequate for non-crossing estimates when measuring estimators for multiple percentiles, resulting in an invalid distribution. On the other hand, the MMQR with fixed effects introduced by Machado and Silva (2019) fully addresses the drawbacks that are associated with panel quantile regression. The conditional quantile QY sjXi,t ðÞ for a location-scale model is expressed as: Yit ¼aiþbX0 it þniþwZ0 it  Uit In this context, Yit and Xit represent the dependent and independently and identically distributed explanatory variables, respectively. Additionally, a,b,n, and ware the coefficients to be estimated. Here, the probability P niþwZ0 it >0  ¼1:Moreover, Z is a k-vector of the known components, while i ¼ 1, ......:, n denotes the individual fixed effects. Uit is independently and identically distributed across individuals i over time t and is orthogonal to Xit, which are standardized to satisfy the moment conditions. Thus, the equation can be rewritten as: QY sjXit ðÞ ¼aiþniqs ðÞðÞ þX0 itbþZ0 itwqs ðÞ Where X0 it is a vector of the independent variable and control variables, aiþdiqs ðÞis the scalar coefficient of the quantile-sfixed or distributional effect at s, which is time invariant. Following Machado and Silva (2019), the MMQR specification of the basic model is as follows: QNCSKEWit sjai,eit,Xi,t ðÞ ¼aiþu1sSENTit þX 5 k¼2 uksCVk,it þeit Table 2. Summary statistics. Variables Obs Mean Std. Dev Min Max NCSKEW 3456 −0.020 0.976 −4.278 3.905 DUVOL 3456 −0.066 1.120 −8.517 4.505 SENT 3456 0.000 1.607 −4.693 4.070 RET 3456 0.005 0.059 −0.449 0.326 SIGMA 3456 0.010 0.006 −0.027 0.070 INF 3456 0.004 0.008 −0.054 0.158 INT 3456 0.041 0.039 −0.001 0.800 Source: Authors. COGENT ECONOMICS & FINANCE 7 investor behavior to react more strongly to news and events, leading to more pronounced fluctuations even during moderate or extreme risk phases. Furthermore, Choi and Yoon (2020) reveal that investor sentiment is significantly shaped by herding behavior during market downturns or extreme conditions, with this herding effect being most pronounced in emerging markets. This trend for investors to follow the crowd during panic market periods exacerbates sentiment-driven fluctuations, making emerging markets particularly susceptible to these fluctuations. As regards frontier markets,Table A7.3 (Supplementary Appendix 7) reveals a significantly negative relationship between investor sentiment and stock market crash risk in the higher quantiles (Q60 to Q90). This indicates that, as sentiment increases, the likelihood of a market crash decreases, particularly in scenarios where crash risk is higher. This finding is particularly insightful for frontier markets, which are characterized by less developed market structures and higher volatility. In these markets, investor sentiment plays an important role in influencing market dynamics, with declines in sentiment significantly increasing the probability of market crashes. Speidell (2009) highlights that individual investors represent a substantial portion of transactions in frontier markets. These investors often favor low-priced stocks, viewing the quantity of shares they can acquire as a measure of value. Additionally, they typically have a higher risk appetite compared to investors in developed markets (Statman, 2008). Their propensity for speculative behavior, or ‘gambler’s’mindset, significantly impacts market stability. During periods of high sentiment, these investors are less likely to exit the market despite apparent risks. Instead, their willingness to embrace higher risks for the potential of greater rewards contributes to market stability and mitigates the risk of crashes. This behavior is particularly relevant at higher quantiles of crash risk (Q70–Q90), where the increased optimism and speculative tendencies among investors help buffer against severe market declines, thus reducing the likelihood of a crash. 6. Conclusions and implications By analysing panel data of 16 stock markets in the Asia-Pacific region through the Method of Moments Quantile Regression (MMQR) approach, this study shows a positive effect of investor sentiment on crash risk at the middle-higher quantiles (Q60–Q90). It also shows that the magnitude and direction of the impact of investor sentiment on crash risk is heterogeneous across levels of market development. To the best of the authors’knowledge, this is the first quantitative study on the influence of investor sentiment on the quantiles of crash risk in stock markets, and therefore this research potentially has important theoretical and practical contributions. In terms of theory, this paper provides support for behavioral theories by giving evidence of a positive effect of investor sentiment on the middle and higher quantiles of stock market crash risk in the Asia-Pacific region. It clarifies the intricate relationship between sentiment and market stability, offering insights into how sentiment influences crash risk differently across developed, emerging, and frontier markets. The findings reveal a positive relationship between investor sentiment and crash risk in higher quantiles for developed and emerging markets, while a negative relationship is observed in frontier markets. Additionally, the investigation of global and local sentiment effects highlights that regional sentiment significantly increases crash risk, particularly in extreme market conditions, while local sentiment generally reduces crash risk in less severe markets. Furthermore, the study constructs investor sentiment indexes using Principal Component Analysis (PCA). As regards practical aspects, the results of this study have implications for global investors, portfolio managers, and policymakers. For global investors, portfolio managers, and policymakers, adapting strategies to the distinct characteristics of developed, emerging, and frontier markets is crucial for mitigating risks and optimizing returns. In developed markets, where market efficiency and regulatory frameworks are more robust, investors should focus on dynamic risk management techniques and incorporate investor sentiment metrics to predict potential market downturns. For emerging markets, which often experience higher volatility and rapid economic changes, portfolio managers should emphasize diversification and stay attuned to regional sentiment shifts, as these can significantly impact market stability. In frontier markets, characterized by their nascent financial systems and higher risk profiles, investors should prioritize thorough local market research and consider the effects of both global and local sentiment on market movements. Policymakers across all market types should enhance transparency, 14 A. T. NGUYEN AND N. T. NGUYEN strengthen financial regulations, and promote investor education to foster market resilience and stability. Tailoring strategies to the specific attributes and risks of each market segment will help in better managing investment risks and achieving more favorable financial outcomes. However, the study has certain limitations. The first stems from not considering the influence of investor sentiment on stock market crash risk while experiencing different states of sentiment (i.e. optimistic/pessimistic). The second concern is that many other determinants affect stock market crash risk but have not been considered in the model. Finally, there is a discrepancy between the number of observations between market groups, which reduces credibility. These gaps are expected to be filled in future studies. Author contributions Methodology: An Tuan Nguyen and Nhung Thi Nguyen; software, An Tuan Nguyen; validation, An Tuan Nguyen and Nhung Thi Nguyen; data curation, An Tuan Nguyen; writing-original draft, An Tuan Nguyen and Nhung Thi Nguyen; writing-review and editing, Nhung Thi Nguyen; visualization, An Tuan Nguyen; supervision, Nhung Thi Nguyen. Disclosure statement No potential conflict of interest was reported by the author(s). Funding This research is funded by Vietnam National Foundation for Science and Technology Development (NAFOSTED) under grant number 502.99-2020.02. About the authors An Tuan Nguyen is a student at the Faculty of Finance and Banking - VNU University of Economics and Business. His research interests are behavioral finance, financial analysis, and financial investment. Assoc. PhD. Nguyen Thi Nhung graduated from the Hanoi Foreign Trade University in Vietnam, and received the Scholarships from French Government for Master and Doctoral Degrees about Finance and Banks at University of Bordeaux in France. She is currently a lecturer, head of department of Investments at the Faculty of Finance and Banking, University of Economics and Business (UEB) –Vietnam National University (VNU). Research fields that she is interested in include financial analysis and investment, risk management, and finance for sustainable development. ORCID An Tuan Nguyen http://orcid.org/0000-0002-0208-8902 Nhung Thi Nguyen http://orcid.org/0000-0002-3648-1964 Data availability statement The data can be made available upon request. References Aboody, D., Even-Tov, O., Lehavy, R., & Trueman, B. (2018). Overnight returns and firm-specific investor sentiment. Journal of Financial and Quantitative Analysis,53(2), 485–505. https://doi.org/10.1017/S0022109017000989 Akarsu, S., & S€ uer, € O. (2022). How investor attention affects stock returns? Some international evidence. Borsa Istanbul Review,22(3), 616–626. https://doi.org/10.1016/j.bir.2021.09.001 Alnafea, M., & Chebbi, K. (2022). Does investor sentiment influence stock price crash risk? Evidence from Saudi Arabia. The Journal of Asian Finance, Economics and Business,9(1), 143–152. An, Z., Chen, Z., Li, D., & Xing, L. (2018). Individualism and stock price crash risk. Journal of International Business Studies,49(9), 1208–1236. https://doi.org/10.1057/s41267-018-0150-z Baker, M., & Wurgler, J. (2006). Investor sentiment and the cross-section of stock returns. The Journal of Finance, 61(4), 1645–1680. https://doi.org/10.1111/j.1540-6261.2006.00885.x COGENT ECONOMICS & FINANCE 15 Baker, M., Wurgler, J., & Yuan, Y. (2012). Global, local, and contagious investor sentiment. Journal of Financial Economics,104(2), 272–287. https://doi.org/10.1016/j.jfineco.2011.11.002 Bashir, U., Kayani, U. N., Khan, S., Polat, A., Hussain, M., & Aysan, A. F. (2024). Investor sentiment and stock price crash risk: The mediating role of analyst herding. Computers in Human Behavior Reports,13, 100371. https://doi. org/10.1016/j.chbr.2024.100371 Blajer-GołeRbiewska, A., Wach, D., & Kos, M. (2018). Financial risk information avoidance. Economic Research,31(1), 521–536. https://doi.org/10.1080/1331677X.2018.1439396 Campbell, J. Y., & Hentschel, L. (1992). No news is good news: An asymmetric model of changing volatility in stock returns. Journal of Financial Economics,31(3), 281–318. https://doi.org/10.1016/0304-405X(92)90037-X Cao, H. H., Coval, J. D., & Hirshleifer, D. (2002). Sidelined investors, trading-generated news, and security returns. Review of Financial Studies,15(2), 615–648. https://doi.org/10.1093/rfs/15.2.615 Chen, J., Hong, H., & Stein, J. C. (2001). Forecasting crashes: Trading volume, past returns, and conditional skewness in stock prices. Journal of Financial Economics,61(3), 345–381. https://doi.org/10.1016/S0304-405X(01)00066-6 Choi, K. H., & Yoon, S. M. (2020). Investor sentiment and herding behavior in the Korean stock market. International Journal of Financial Studies,8(2), 34. https://doi.org/10.3390/ijfs8020034 Chui, A. C., Titman, S., & Wei, K. J. (2010). Individualism and momentum around the world. The Journal of Finance, 65(1), 361–392. https://doi.org/10.1111/j.1540-6261.2009.01532.x Cui, H., & Zhang, Y. (2020). Does investor sentiment affect stock price crash risk? Applied Economics Letters,27(7), 564–568. https://doi.org/10.1080/13504851.2019.1643448 De Long, J. B., Shleifer, A., Summers, L. H., & Waldmann, R. J. (1990). Noise trader risk in financial markets. Journal of Political Economy,98(4), 703–738. https://doi.org/10.1086/261703 Dumitrescu, E. I., & Hurlin, C. (2012). Testing for Granger non-causality in heterogeneous panels. Economic Modelling, 29(4), 1450–1460. https://doi.org/10.1016/j.econmod.2012.02.014 Fama, E. F. (1970). Efficient capital markets. Journal of Finance,25(2), 383–417. Fan, Y., Zhou, F., An, Y., & Yang, J. (2021). Investor sentiment and stock price crash risk: Evidence from China. Global Economic Review,50(4), 310–339. https://doi.org/10.1080/1226508X.2021.1947340 French, K. R., Schwert, G. W., & Stambaugh, R. F. (1987). Expected stock returns and volatility. Journal of Financial Economics,19(1), 3–29. https://doi.org/10.1016/0304-405X(87)90026-2 Fu, J., Wu, X., Liu, Y., & Chen, R. (2021). Firm-specific investor sentiment and stock price crash risk. Finance Research Letters,38, 101442. https://doi.org/10.1016/j.frl.2020.101442 Fu, J., Zhou, Q., Liu, Y., & Wu, X. (2020). Predicting stock market crises using daily stock market valuation and investor sentiment indicators. The North American Journal of Economics and Finance,51, 100905. https://doi.org/ 10.1016/j.najef.2019.01.002 Habib, A., Hasan, M. M., & Jiang, H. (2018). Stock price crash risk: review of the empirical literature. Accounting & Finance,58(S1), 211–251. https://doi.org/10.1111/acfi.12278 Hong, H., & Stein, J. C. (2003). Differences of opinion, short-sales constraints, and market crashes. Review of Financial Studies,16(2), 487–525. https://doi.org/10.1093/rfs/hhg006 Hutton, A. P., Marcus, A. J., & Tehranian, H. (2009). Opaque financial reports, R2, and crash risk. Journal of Financial Economics,94(1), 67–86. https://doi.org/10.1016/j.jfineco.2008.10.003 Jang, J., & Kang, J. (2019). Probability of price crashes, rational speculative bubbles, and the cross-section of stock returns. Journal of Financial Economics,132(1), 222–247. https://doi.org/10.1016/j.jfineco.2018.10.005 Jiang, M., & Zhu, B. (2023). The nonlinear relationship between investor sentiment and stock crash risk under the regime of EPU-evidence from China’s individual stock and market. Available at SSRN 4439527. Kao, C., & Chiang, M. H. (2001). On the estimation and inference of a cointegrated regression in panel data. In Nonstationary panels, panel cointegration, and dynamic panels (pp. 179–222). Emerald Group Publishing Limited. Kim, Y., Li, H., & Li, S. (2014). Corporate social responsibility and stock price crash risk. Journal of Banking & Finance, 43,1–13. https://doi.org/10.1016/j.jbankfin.2014.02.013 Kothari, S. P., Shu, S., & Wysocki, P. D. (2009). Do managers withhold bad news? Journal of Accounting Research, 47(1), 241–276. https://doi.org/10.1111/j.1475-679X.2008.00318.x Kustina, L., Sudarsono, R., & Effendi, N. (2024). Does foreign portfolio investment moderate the impact of exchange rate volatility and investor sentiment on country index crash risk? Cogent Economics & Finance,12(1), 2305481. https://doi.org/10.1080/23322039.2024.2305481 Liu, Z., Huynh, T. L. D., & Dai, P. F. (2021). The impact of COVID-19 on the stock market crash risk in China. Research in International Business and Finance,57, 101419. https://doi.org/10.1016/j.ribaf.2021.101419 Lucey, B. M., & Dowling, M. (2005). The role of feelings in investor decision-making. Journal of Economic Surveys, 19(2), 211–237. https://doi.org/10.1111/j.0950-0804.2005.00245.x Machado, J. A., & Silva, J. S. (2019). Quantiles via moments. Journal of Econometrics,213(1), 145–173. https://doi.org/ 10.1016/j.jeconom.2019.04.009 Mishkin, F. S., & White, E. N. (2002). US stock market crashes and their aftermath: Implications for monetary policy (NBER Working Paper 8992). https://doi.org/10.7551/mitpress/1459.003.0011 16 A. T. NGUYEN AND N. T. NGUYEN Moradi, M., Appolloni, A., Zimon, G., Tarighi, H., & Kamali, M. (2021). Macroeconomic factors and stock price crash risk: Do managers withhold bad news in the crisis-ridden Iran market? Sustainability,13(7), 3688. https://doi.org/ 10.3390/su13073688 Patel, S. A., & Sarkar, A. (1998). Crises in developed and emerging stock markets. Financial Analysts Journal,54(6), 50–61. https://doi.org/10.2469/faj.v54.n6.2225 Peng, L., & Xiong, W. (2006). Investor attention, overconfidence and category learning. Journal of Financial Economics,80(3), 563–602. https://doi.org/10.1016/j.jfineco.2005.05.003 Phillips, P. C., & Hansen, B. E. (1990). Statistical inference in instrumental variables regression with I (1) processes. The Review of Economic Studies,57(1), 99–125. https://doi.org/10.2307/2297545 Ross, S. A. (1973). The economic theory of agency: The principal’s problem. The American Economic Review,63(2), 134–139. Shin, H., & Choi, S. Y. (2022). Firm-specific investor sentiment and stock price crash risk: The role of foreign investors in Korea’s stock market. Investment Management and Financial Innovations,19(4), 309–326. https://doi.org/10. 21511/imfi.19(4).2022.25 Speidell, L. S. (2009). Investing in the unknown and the unknowable—Behavioral finance in frontier markets. Journal of Behavioral Finance,10(1), 1–8. https://doi.org/10.1080/15427560902719323 Statman, M. (2008). Countries and culture in behavioral finance. CFA Institute Conference Proceedings Quarterly,25(3), 38–44. https://doi.org/10.2469/cp.v25.n3.6 Vo, X. V. (2020). Foreign investors and stock price crash risk: Evidence from Vietnam. International Review of Finance, 20(4), 993–1004. https://doi.org/10.1111/irfi.12248 Vuong, N. B., & Suzuki, Y. (2020). Impact of financial development on sentiment-return relationship: Insight from Asia-Pacific markets. Borsa Istanbul Review,20(2), 95–107. https://doi.org/10.1016/j.bir.2020.01.003 Wang, J., Meric, G., Liu, Z., & Meric, I. (2009). Stock market crashes, firm characteristics, and stock returns. Journal of Banking & Finance,33(9), 1563–1574. https://doi.org/10.1016/j.jbankfin.2009.03.002 Westerlund, J. (2007). Testing for error correction in panel data. Oxford Bulletin of Economics and Statistics,69(6), 709–748. https://doi.org/10.1111/j.1468-0084.2007.00477.x Wu, B., Cai, Y., & Zhang, M. (2021). Investor sentiment and stock price crash risk in the Chinese stock market. Journal of Mathematics,2021,1–10. https://doi.org/10.1155/2021/6806304 Yang, C., & Gao, B. (2014). The term structure of sentiment effect in stock index futures market. The North American Journal of Economics and Finance,30, 171–182. https://doi.org/10.1016/j.najef.2014.09.001 Yin, Y., & Tian, R. (2017). Investor sentiment, financial report quality and stock price crash risk: Role of short-sales constraints. Emerging Markets Finance and Trade,53(3), 493–510. https://doi.org/10.1080/1540496X.2015.1093844 Zhang, P., Gao, J., Zhang, Y., & Wang, T. W. (2021). Dynamic spillover effects between the US stock volatility and China’s stock market crash risk: a TVP-VAR approach. Mathematical Problems in Engineering,2021,1–12. https:// doi.org/10.1155/2021/6616577 Zhang, R., Xian, X., & Fang, H. (2019). The early-warning system of stock market crises with investor sentiment: Evidence from China. International Journal of Finance & Economics,24(1), 361–369. https://doi.org/10.1002/ijfe.1667 Zhang, Z., Chen, R., & Luo, Q. (2023). Firm-specific investor sentiment, stock price synchronicity, and crash risk. Applied Economics Letters,30(4), 450–455. https://doi.org/10.1080/13504851.2021.1991562 Zhou, G. (2018). Measuring investor sentiment. Annual Review of Financial Economics,10(1), 239–259. https://doi.org/ 10.1146/annurev-financial-110217-022725 Zhou, L., Chen, D., & Huang, J. (2023). Stock-level sentiment contagion and the cross-section of stock returns. The North American Journal of Economics and Finance,68, 101966. https://doi.org/10.1016/j.najef.2023.101966 Zhou, W., Zhong, G. Y., & Li, J. C. (2022). Stability of financial market driven by information delay and liquidity in delay agent-based model. Physica A: Statistical Mechanics and Its Applications,600, 127526. https://doi.org/10. 1016/j.physa.2022.127526 Zouaoui, M., Nouyrigat, G., & Beer, F. (2011). How does investor sentiment affect stock market crises? Evidence from panel data. Financial Review,46(4), 723–747. https://doi.org/10.1111/j.1540-6288.2011.00318.x COGENT ECONOMICS & FINANCE 17