DOI: 10.28934/ea.22.55.2.pp20-30 First Online: November 1, 2022 ORIGINAL SCIENTIFIC PAPER Covid-19 and Serbian Stock Market Response: A Panel Data Approach Bojan Đorđević1 | Sunčica Stanković14F* 1 Megatrend university Belgrade, Faculty of Management Zaječar, Zaječar, Republic of Serbia ABSTRACT In this paper, the authors attempted to explore the relationship between the Covid-19 Coronavirus pandemic and the stock market in the Republic of Serbia. The main research variables on the stock market are the daily values of the Belex 15 and Belex Line indices and trading volume. For the pandemic variables, official daily data on the number of new Covid-19 cases in Serbia, Europe, and the world were taken. By applying the panel regression analysis for the period from 03/06/2020 to 12/30/2021, the empirical results show a positive and significant influence of the number of daily infected in Serbia and Europe on the stock market index value, At the same time, the influence of daily infected ones on trading volume is negative but statistically significant only when it comes to the new cases of coronavirus per day at the level of Europe. The presented results indicate the resilience of the Serbian capital market to internal and external shocks. Keywords: Covid-19, Stock market indices, Belgrade Stock Exchange, Panel data analysis JEL Classification: C33, D53, I10 INTRODUCTION The Covid-19 pandemic, besides the uncertainty and panic, led to a temporary lockdown in many states and the slowdown of economic activities. Almost two years after the pandemic started, the companies’ performance downfall is visible in all sectors, affecting the financial markets, specifically stock markets (Okorie and Lin, 2021). According to Zaremba et al. (2020) and Singh et al. (2020), in terms of the Covid-19 crisis, the intervention of governments and central banks of numerous countries has significantly increased the volatility of the international capital market. At the same time, the official statistics of WHO mark the increase in the daily number of infections and deaths all over the world, along with the growing resistance of the population towards vaccination (World Health Organization, 2021). Financial markets reacted quickly, reflecting the magnitude of the crisis potential. In March 2020, the S&P 500, Euro Stoxx 600, and Nikkei 225 indices plummeted by 9.5, 8.3, and over 10%, respectively. A similar situation happened with the Chinese financial market, affecting the value of the SSE Composite Index of 7.72% and the SZSE Component Index of 8.45%. The indices of the Belgrade stock market on the Serbian capital market have experienced, during February and March 2020, a significant drop in their values, following global stock indices. Companies’ stock prices and Belex 15 and Belex Line index values continually and intensively kept plunging until March 24, 2020. Since then, the values have been going up. The initial * Corresponding author, e-mail: suncica.stankovic@fmz.edu.rs Bojan Đorđević, Sunčica Stanković 21 reaction of markets was brutal and escalated, which is commonly expected investors' behavior in times like these. On the other hand, there is an opinion that the market recovered by investors’ faith in normalization time and life and economic activities (Naseem et al., 2021). Crucial psychological moments and positive signs for investors, which also transmitted to the Serbian capital market, came from the global vaccine manufacturers – first in the form of development announcements and later as the presence and distribution of Covid-19 vaccines. Bearing in mind the current state of the pandemic and the dynamic of price movements on global stock markets, the subject of this research is the Covid-19 and Serbian stock market response through the implementation of the panel regression analysis. The goal is to investigate relations between indices values and trading volume on the Belgrade stock market on one side, and daily infections in Serbia, Europe, and the world, on the other. Following the subject and goal of the research, there are two questions: 1) Does the number of daily infected in Serbia, Europe, and the world affect the value of the Belgrade stock market index? 2) Does the number of daily infected in Serbia, Europe, and the world affect the volume of trading on the Belgrade stock market? To answer the research questions, we structured the paper as follows: after the introduction, the second part gives a short literature review, reflecting on the most significant results; the third part presents research methods; part four exposes empirical results, while the last part portrays conclusions, restrictions, and recommendations for future studies. LITERATURE REVIEW The global pandemic contributed to an increase in research on Covid-19 influence on financial markets. Analyzing key indicators of pandemic effects on overall and local levels (the number of infected, deceased, and vaccinated, in correlation with prices and yield for most relevant global and regional stock market indices), most of the obtained results show a significant correlation and the effect of a current pandemic on market volatility and stock price movements. A short review of some latest research is set hereafter. Singh et al. (2020) examined the pandemic impact on the stock markets of G-20 countries, using an event study to measure abnormal returns (ARs) and panel data regression to explain the causes of ARs. The observed window comprises 58 days after the news of the Covid-19 outbreak was released on the international media, and the estimation window consists of 150 days before the announcement. The results confirmed the recovery of stock markets from the Covid-19 negative impact. Zaren and Hizarci (2020) analyzed the possible effects of Covid-19 on stock markets using daily stock indices data. The cointegration test using Covid-19 daily infections and deaths was used to question possible outcomes on the stock markets. The SSE, KOSPI, and IBEX35 indices have a cointegration relationship with the number of infections, while FTSE MIB, CAC40, DAX30 indices don't. Onali (2020) researched the effects of the pandemic on the Dow Jones and S&P500 indices. The outcomes confirmed the Covid-19 crisis did not affect the US stock market returns. However, VAR models indicated the number of coronavirus-related deaths in Italy and France had negative implications on stock market returns and positive ones on the VIX returns. Ali et al. (2020) tested the effect of the Covid-19 crisis on the global financial market. Analyzing periods of epidemic and pandemic, they stress the substantial effects of the pandemic on commodity and stock markets. The results show a negative correlation between gold price movements and the resistance of the Chinese capital market. Baker et al. (2020) gained results of the pandemic’s severe effect on the US stock market compared to the 20th-century historical crises. Höhler and Lansink (2020) measured the pandemic influence on the volatility of stocks of companies that manufacture and distribute consumer goods listed within the most relevant global stock market indices. The results show considerable effects of the pandemic on the growth of volatility of all companies’ stocks, apart from the ones of companies that produce and supply food. Bai et al. (2020) used the GARCH-MIDAS model to analyze the pandemic effect on stock markets in the USA, China, Great Britain, and Japan. A significant impact on the international stock market was 22 Economic Analysis (2022, Vol. 55, No. 2, 20-30) evident, but individually a small effect on the Chinese stock market. Lee and Lu (2021) researched how the Covid-19 pandemic affected the Taiwanese stock market. The result was the stocks of companies with a higher level of corporate social responsibility (CSR) are more resilient to the current crisis. Ozkan (2021) tested a market efficiency hypothesis on the stock market of several developed countries. The gained results of econometric models indicate high power to predict abnormal yields during the pandemic. Hsu and Liao (2022) explored behavior in the US market, with stock volatility, trading volume, and yield as elementary variables on one side and the number of infected and deceased on the other side. There is a positive correlation between the coronavirus effect and stock price volatility and trading volume and a negative correlation between the pandemic and stock yield. Amin et al. (2021) inspected the pandemic effect on stock market indices in North, Central, and South America. Through panel regression analyses, they deducted the number of infected has a considerable bad influence on stock price volatility, except for the South American countries, with no significant statistical correlation. Sahoo (2021) and Bora and Basistha (2021) studied the Covid-19 influence on the Indian stock market, comparing stock index yields from two different periods – before the crisis and during one. The outcomes obtained by the regressional and GARCH model show a statistically significant positive correlation with stock market volatility and movements during the pandemic. Tapaloglu et al. (2021) used the panel data analysis method to present the relationship between the pandemic and stock markets in Turkey, Belgium, Germany, France, Italy, Spain, the United Kingdom, the United States, China, and the Netherlands. Covid-19 data is based on the total number of cases and the total number of deaths, while stock market data relies on major stock indices of countries. There was a negative relationship between the total number of cases and the stock market and a positive one between the total number of deaths and the stock market. Khalid et al. (2021) used a panel quantile regression model of 17 developed stock markets to show the pandemic's impact on stock market returns and volatility. They proved there is no significant impact of the coronavirus on stock returns. Moreover, it had a positive effect on stock market volatility. RESEARCH METHODOLOGY To analyze the relation between daily infections in Serbia, Europe, and the world and index value on the Belgrade stock market, that is, trading volume, we used daily data for the period 03/06/2020 to 12/30/2021. There is a time series (T) of 462 days, while the number of observed entities (N) is two (Belex Line and Belex 15). Therefore, the number of observations encompassed within these panel analyses is 924. Due to the robustness of the data, the dependent variables are logarithmically transformed. Data on daily infected was gathered from the web portal Our World in Data, while the information on index value and trading volume was from the Belgrade Stock Exchange web page. Table 1. Description of the Variables Used in the Regression Analysis Variable Description Source BELEX 15 Indice - B15 Stocks of the 11 most liquid Serbian companies - Daily data of B15 value Belgrade Stock Exchange https://www.belex.rs/trgovanje/indek si/belex15/istorijski BELEX Line Indice - BL Stocks of the 34 Serbian companies - Daily data of BL value Belgrade Stock Exchange https://www.belex.rs/trgovanje/indek si/belexline/istorijski Trading volume Daily data of the number of stocks Belgrade Stock Exchange https://www.belex.rs/trgovanje/indek si/belex15/istorijski https://www.belex.rs/trgovanje/indek si/belexline/istorijski Bojan Đorđević, Sunčica Stanković 23 Variable Description Source Number of new cases of Coronavirus in SERBIA Daily data WHO https://covid19.who.int/ Number of new cases of Coronavirus in Europe Daily data WHO https://covid19.who.int/ Number of new cases of Coronavirus in WORLD Daily data WHO https://covid19.who.int/ The dependent variables of the research are as follows: the first analysis – the Belgrade Stock Exchange index value (Belex 15 and Belex Line); the second analysis – tk trading volume. The independent variables in both analyses are the number of daily infections in Serbia, Europe, and the world. The description of variables is presented in Table 1. The descriptive statistics of variables used in the analyses are given in Table 2. It shows calculated values of the central tendency measures and variabilities. Data on the number of observations, arithmetic mean, standard deviations, that is, the average deviation of arithmetic mean, minimal and maximal parameter values are in the columns. Table 2. Descriptive statistics of the research variable Variable Obs. Mean Std. Dev Min Max Stock Indices values 924 1156.358 426.6729 606.62 1721.16 Trading volume 924 1.08e+07 3.42e+07 26131 4.52e+08 Serbia 924 2081.121 2438.846 0 9983 Europe 924 156306 128847.1 1510 989061 World 924 540773.1 176226.8 25521 1934140 Source: Authors’ calculation The following graphs (Figure 1 and Figure 2) illustrate the movement of Belgrade stock market index value, Belex Line and Belex 15, and trading volume from March 6, 2020, to December 30, 2021. Based on Figure 1, it can be noted that the movement trend of both indices is uniform. After the start of the Coronavirus pandemic, the value of both indices decreased, and then with the stabilization of the financial market, the value of both indices increased. Živković (2022) also points to the marked volatility and uniform trend of the observed indices in the period from 2008-2022. Figure 1. Movement of stock market index values and trading volume on BSE Source: Authors 50 0 10 00 15 00 20 00 mar2020 dec2021mar2020 dec2021 Belex line Belex 15 In de x va lu es 0 5. 00 0e +0 8 mar2020 dec2021mar2020 dec2021 Belex line Belex 15 Tr ad in g vo lu m e 24 Economic Analysis (2022, Vol. 55, No. 2, 20-30) RESEARCH METHODS Research methodology and data analyses are based on panel data, that is, on regression models of panel data (Panel Data Regression Model – PDRM). The preliminary part of econometric analysis evaluates different formulations of statistic models and then runs various tests to choose the most suitable model for research data, as well as econometrical diagnostic tests to check if the model presumptions are fulfilled (specification model errors, multicollinearity, auto-correlation, and heteroscedasticity). The analysis uses strictly balanced datasets (“full” time series). The least-squares model (Pooled OLS – POLS), fixed-effect model (FE), and random-effect model (RE) were used for testing. The following model is specified to explain the dependent variables by using independent ones: 𝑌𝑌𝑚𝑚𝑡𝑡 = 𝛼𝛼 + 𝛽𝛽1𝑋𝑋1,𝑚𝑚𝑡𝑡 + 𝛽𝛽2𝑋𝑋2,𝑚𝑚𝑡𝑡 + 𝛽𝛽3𝑋𝑋3,𝑚𝑚𝑡𝑡 + 𝑢𝑢𝑚𝑚 + 𝑒𝑒𝑚𝑚𝑡𝑡 (1) Where: 𝑌𝑌𝑚𝑚𝑡𝑡 – dependent variable: in the first research ln(index value – IV), in the second research ln(trading volume – TV). 𝑖𝑖 – entity (in the first analysis: 1 = Belex Line value, and 2 = Belex 15 value; in the second analysis: 1= Belex Line Trading volume, and 2= Belex 15 Trading volume); 𝑡𝑡 – time (in both analyses: 1 = 06.03.2020.... 462 = 31.12.2021.). 𝛽𝛽 – coefficient for respective independent variables; 𝛼𝛼 – intercept; 𝑋𝑋1 – independent variable (number of daily infections in Serbia - S); 𝑋𝑋2– independent variable (number of daily infections in Evropi - E); 𝑋𝑋3 – independent variable (number of daily infections in the world - W); 𝑢𝑢𝑚𝑚 – the individual impact of the ith entity; 𝑒𝑒𝑚𝑚𝑡𝑡 – the error term. RESULTS OF ECONOMETRIC TESTS a) Model 1 was evaluated through econometric tests, where the dependent variable of the stock market index value (IV) is given as a linear stochastic function of independent, i.e., explanatory variables: 𝐼𝐼𝐼𝐼𝑚𝑚𝑡𝑡 = 𝛼𝛼𝑚𝑚 + 𝛽𝛽1𝑆𝑆 + 𝛽𝛽2𝐸𝐸 + 𝛽𝛽3𝑊𝑊 + 𝑢𝑢𝑚𝑚 + 𝑒𝑒𝑚𝑚𝑡𝑡 (2) The random effects model was chosen as the most appropriate model (Table 4). According to the determination coefficient, for the RE model, 9% (R2 = 0.09) variations of the dependent variable are explained based on independent variables. The values of F statistics point to statistical significance RE model (Prob > F = 0.0000). Hausman’s test checked whether the individual effects are correlated with regressors, in which case the FE model would be more suitable. According to Hausman’s test results, the significance is higher than 0.05 (χ2 = 0.00, p = 1.0000). Thus, we accept the null hypothesis claiming that the RE model is more appropriate than the FE model. The justification of the RE model is tested with the Breusch and Pagan LM test (Lagrange multiplier) for testing the existence of individual effects. If the test significance is > 0.05, the null hypothesis (no heterogeneity between observed entities) cannot be rejected, which means POLS is more suitable than the RE model. The test results indicate the level of significance is lower than the set level (χ2 = 2.0e+05, p = 0.0000), so we can conclude the RE model is more suitable than the POLS model. Ramsey RESET test was used to check if the model Bojan Đorđević, Sunčica Stanković 25 is well-specified. The results gained (F(3, 917) = 0.62; Prob > F = 0.6049) indicate no significant variables left out of the model. In the next step, with the use of the Pasaran CD test, we looked for serial correlation problems in the model. The null hypothesis is – no serial correlation. The statistical significance of the test is higher than 0.05 (p = 0.4511), meaning the null hypothesis about the non-existence of a serial correlation can be accepted. Also, Wooldridge’s test for autocorrelation in the panel data indicates there is no serial correlation (p = 0.1910). White’s test was used for heteroscedasticity testing in model 1. If the χ2 statistic probability obtained by this test is higher than the error risk α (α = 5%), the null hypothesis cannot be rejected. The probability value of chi statistics in this test is 0.1121, and with the error risk of 5%, we cannot discard the null hypothesis, confirming the homoscedasticity errors in the model. Results from the panel analysis for the RE model indicate a positive and significant influence of the number of daily infections in Serbia and Europe on the index value of the Belgrade stock market, while the effect on a global level is positive but not statistically significant. Table 4. Evaluation of random-effect model with index value dependent variable, and diagnostic tests Index values Coef. Std. Err. t P>|t| Serbia 5.60e-06 8.97e-07 6.24 0.0001 Europe 2.16e-07 1.79e-08 12.10 0.0001 World 1.15e-08 1.15e-08 0.99 0.321 Constant 6.928381 .3817168 18.15 0.0001 Diagnostic tests Hausman’s test χ2(3) = 0.00; p = 1.0000 LM test chibar2(01) = 2.0e+05; Prob > chibar2 = 0.0000 Ramsey RESET test F(3, 917) = 0.62; Prob > F = 0.6049 Pasaran CD test 0.438; p = 0.4511 Wooldridge’s test F(1, 1) = 18.593; Prob > F = 0.1910 White’s test χ2 (9) = 12.60; Prob > χ2 = 0.1121 Note: R squere = 0.09; Prob > F = 0.0000; Root MSE = 0.39; Number of observations = 924. Source: Authors’ calculation To test linear dependence between explanatory variables, that is, to probe the existence of harmful multicollinearity in a regression model, the variance inflation factor (VIF), as well as tolerance factor – 1/VIF, were used. Based on the results in Table 5, it can conclude the model has no harmful multicollinearity. Table 5. Multicollinearity test Variable VIF 1/VIF Serbia 1.56 0.640600 Europe 1.73 0.577339 World 1.35 0.740278 Mean VIF 1.55 Source: Authors’ calculation b) Model 2 was econometrically tested, where the dependent variable stock trading volume (TV) is presented as a linear stochastic function of independent variables: 𝑇𝑇𝐼𝐼𝑚𝑚𝑡𝑡 = 𝛼𝛼𝑚𝑚 + 𝛽𝛽1𝑆𝑆 + 𝛽𝛽2𝐸𝐸 + 𝛽𝛽3𝑊𝑊 + 𝑢𝑢𝑚𝑚 + 𝑒𝑒𝑚𝑚𝑡𝑡 (3) 26 Economic Analysis (2022, Vol. 55, No. 2, 20-30) Table 6. Evaluation of RE model with stock trading volume-dependent variable, and diagnostic tests Trading volume Coef. Std. Err. t P>|t| Serbia -.0000211 .000022 -0.96 0.338 Europe -8.83e-07 4.38e-07 -2.01 0.044 World -6.48e-08 2.83e-07 -0.23 0.819 Constant 15.40123 .1904107 80.88 0.0001 Diagnostic tests Hausman’s test χ2(3) = 0.00, p = 1.0000 LM test chibar2(01) = 8.50; Prob > chibar2 = 0.0018 Ramsey RESET test F(3, 917) = 0.60; Prob > F = 0.5913 Pasaran CD test 0.638; p = 0.3312 Wooldridge’s test F(1, 1) = 17.904; Prob > F = 0.1477 White’s test chi2(9) = 13.20; Prob > chi2 = 0.1536 Note: R squere = 0.13; Prob > F = 0.007; Number of groups = 2; Number of obs. = 924. Source: Authors’ calculation The random effects model was chosen as the most appropriate model (Table 6). As stated in Table 6, parameters with the independent variable are statistically significant. The coefficient determination value denotes that 13% of the dependent variable variations (stock trading volume) are explained with the RE model, and the F statistic values suggest statistical significance in the RE model (Prob > F = 0.007). By Hausman test results, the significance is higher than 0.05 (χ2 = 0.00, p = 1.000), so we accept the null hypothesis the RE model is more suitable than the FE model. The Breusch-Pagan LM test results point out that the significance level is lower than the set one (χ2 = 8.50, p = 0.0018), so the conclusion is the RE model is more suitable compared to POLS. The results from the Ramsey RESET test (Prob > F = 0. 5913) prove there are no significant variables left out of the model, meaning the model specification is good. Pasaran test statistical significance is higher than 0.05 (p = 0.3312), so the null hypothesis on serial correlation non-existence is acceptable. Wooldridge’s test for autocorrelation in the panel data indicates there is no serial correlation (p = 0.1477). White’s test was used to test the model’s heteroskedasticity. The value of χ2 statistics probability in this test is 0.1536, which confirms the model errors are homoskedastic. According to the results for the RE model, trading volume on the Belgrade stock market is negative but not statistically significant for the number of daily infections in Serbia and the world, while the number of newly infected in Europe is negative on a statistically significant level. CONCLUSION The conducted research had a goal to probe the correlation between the index stock value of Belex15 and Belex Line, trading volume on the Belgrade stock market on one side, and the number of newly infected in Serbia, Europe, and the world, on the other side. The results gained by the panel regression analysis for the random effect model (RE), which was econometrically tested, well-specified, and most appropriate for the research, indicate a positive and significant influence of daily infections in Serbia and Europe on the Belgrade stock market index value, while the global effect is positive but not statistically significant (model 1). These results are unexpected. However, soon after the beginning of the pandemic, when the value of both indices fell, the government took preventive measures, with the aim of maintaining economic stability, by supporting micro, small and medium enterprises. This package of economic measures to Bojan Đorđević, Sunčica Stanković 27 mitigate the consequences of the coronavirus included: tax policy measures (such as, for example, deferring the payment of payroll taxes), direct assistance to companies for employees (e.g., direct assistance to entrepreneurs who are taxed at a flat rate and who pay income tax real income, to micro, small and medium-sized enterprises in the private sector - payment of assistance in the amount of the minimum), measures for liquidity of the economy (e.g. support to the economy through the Development Fund of the Republic of Serbia) and other measures (e.g. payment of 100 euros to all citizens of legal age). These preventive measures obviously had a positive impact on the Serbian capital market. The obtained results are in accordance with the research results of Waheed et al. (2020), who examined the impact of the covid pandemic on the Karachi stock exchange and concluded that the covid pandemic has a diverse impact on developing economies, compared to developed economies, which faced serious declines in this period. In the second analysis (model 2), econometric tests also proved it is best to use the model with random effect (RE). The results gained by the panel regression analysis for the random effect model (RE) indicate a negative but statistically insignificant correlation between the number of newly infected in Serbia and the world and the trading volume on the Belgrade stock market. The number of newly infected in Europe correlates negatively with the trading volume on the Belgrade stock market on a statistically significant level. The obtained results are consistent with the results of research on other stock markets, where we refer to Öztürk et al. (2020), Zaren and Hizarci (2020), Onali (2020), Đorđević and Stanković (2021), Naseem et al. (2021), who have empirically gained similar conclusions. According to the research results, R square records small values for both models (9% for the first and 13% for the second model). This means that only a small percentage of the variations of the dependent variable (in the first model stock market index value, and in the second stock trading volume) can be explained by means of independent variables (the number of daily infected in Serbia, Europe, and the world). This speaks in favor of the fact that variations in the capital market in Serbia depend on other important explanatory variables, which should be included in the model. However, according to the subject of the research, only Covid 19 variables were included in the research, and it can be concluded that these variables can explain a very small part of the variations in the Serbian stock market. Historically speaking, after the initial panic and significant fall of the Belex 15 stock index (fall of over 200 index points) in February/March 2020, the Serbian stock market recovered and somewhat adapted to the uncertainty caused by the pandemic. In 2021, there were two strategic moves to develop the capital market in Serbia: 1. Consolidation of the Commercial Bank (listed on The Belgrade Stock Exchange: KMBN) by the Slovenian NLB Group, and 2. The strategic partnership of the Belgrade and Athens stock market (Greece) was one of the ways to improve the capital market and set the stock market as a central piece of the Serbian economy. This move would contribute to better visibility and attractiveness of home market securities to foreign investors who are present and trade in inconsiderable volume, not accounting for the current conditions (BBC News, 2021). We can conclude that, under the current turbulent conditions, there has been a quick recovery of the stock market, there was no greater capital outflow from the Belgrade stock market, and the shock hasn’t been enormous and devastating. Simultaneously, in 2020/2021, all other significant stock indices in Europe and the world recorded a swift recovery and growth, a characteristic of stock markets. Favorable information gained in 2020 from the Covid-19 vaccination manufacturers (Phizer, Astra Zeneca), and a set of economic measures of many countries, contributed to a better investors’ psychological climate in all markets. In a crisis, panic in the market is not in compliance with long-term investment strategies. The current pandemic is not only a possibility but also a necessity for investors to consider their investment portfolios and carry out a hedge and optimize. Many authors and stock market analysts highlight the inclusion of gold as a safe asset, futures and options trade, 28 Economic Analysis (2022, Vol. 55, No. 2, 20-30) and also the inclusion of some cryptocurrencies into portfolios for diversification and reduction of risk to a minimum. The limitations of this research are in its sensitivity to new information and data. For that, we need further research that would include new information and data on the subject, like stock yield and trading analysis, number of deaths, number of vaccinated people in Serbia, Europe, and on the global level, to secure valid information for politicians, investors, portfolio managers and CEOs in the decision-making process. The crucial implication of this research is a further study on factors connected with pandemic waves that affect the volatility of the Serbian stock market. In the end, the results of this study will be of use to the domestic and foreign professional public because they will provide valuable information on the behavior of financial markets in developing countries during the crisis. Also, the study can be helpful to state decision-makers to implement the necessary activities and reduce the potential undesired effects of the current health crisis. REFERENCES Abuzayed, Bana, Elie Bouri, Nedal Al-Fayoumi, and Jalkh Naji. 2021. “Systemic risk spillover across global and country stock markets during the COVID-19 pandemic.” Economic Analysis and Policy, 71: 180–197. https://doi.org/10.1016/j.eap.2021.04.010 Adcock, Christopher, Xiuping Hua, Khelifa Mazouz, and Shuxing Yin. 2014. “Does the stock market reward innovation? European stock index reaction to negative news during the global financial crisis.” Journal of International Money and Finance, 49(PB): 470–491. https://doi.org/10.1016/j.jimonfin.2014.06.004 Agosto, Arianna, Daniel F. Ahelegbey, and Paolo Giudici. 2020. “Tree networks to assess financial contagion.” Economic Modelling, 85: 349–366. https://doi.org/10.1016/j.econmod.2019.11.005 Ali, Mohsin, Nafis Alam, and Syed Aun R. Rizvi. 2020. “Coronavirus (COVID-19) — An epidemic or pandemic for financial markets.” Journal of Behavioral and Experimental Finance, 27: 100341. https://doi.org/10.1016/j.jbef.2020.100341 Amin, Ali Muhammad Arshad, Naheed Sultana, Rabeeya Raoof. 2021. “Examination of impact of COVID-19 on stock market: evidence from American peninsula.” Journal of Economic and Administrative Sciences, Vol. ahead-of-print No. ahead-of-print. https://doi.org/10.1108/JEAS-07-2020-0127 Andersen, Torben G., Tim Bollerslev, Francis X. Diebold, And Heiko Ebens. 2001. „The Distribution of Realized Stock Return Volatility.” Journal of Financial Economics, 61(1): 43–76. http://dx.doi.org/10.1016/S0304-405X(01)00055-1 AZIMLI, Asil. 2020. “The impact of COVID-19 on the degree of dependence and structure of risk- return relationship: a quantile regression approach.” Finance Research Letters, 36(C) https://doi.org/https://doi.org/10.1016/j.frl.2020.10 Bai, Lan, Yu Wei, Guiwu Wei, Xiafei Li, and Songyun Zhang. 2021. „Infectious disease pandemic and permanent volatility of international stock markets: A long-term perspective.” Finance Research Letters, 40: 1-10. https://doi.org/10.1016/j.frl.2020.101709 Baker, Scott R., Nicholas Bloom, Steven J. Davis, Kyle J. Kost, Marco C. Sammon, and Tasaneeya Viratyosin. 2020. “The unprecedented stock market reaction to COVID-19.” Working Paper 26945. Cambridge: National Bureau of Economic Research. https://www.nber.org/system/files/working_papers/w26945/w26945.pdf (accessed June 22, 2022). BBC NEWS. 2021. Tržište kapitala u Srbiji: Šta radi i čemu služi Beogradska berza. https://www.danas.rs/bbc-news-serbian/trziste-kapitala-u-srbiji-sta-radi-i-cemu-sluzi- beogradska-berza/ (accessed April 20, 2022). Belgrade Stock Exchange. 2022. Data. Available at: https://www.belex.rs https://doi.org/10.1016/j.eap.2021.04.010 https://doi.org/10.1016/j.jimonfin.2014.06.004 https://doi.org/10.1016/j.econmod.2019.11.005 https://doi.org/10.1016/j.jbef.2020.100341 https://doi.org/10.1108/JEAS-07-2020-0127 http://dx.doi.org/10.1016/S0304-405X(01)00055-1 https://doi.org/https:/doi.org/10.1016/j.frl.2020.10 https://doi.org/10.1016/j.frl.2020.101709 https://www.nber.org/system/files/working_papers/w26945/w26945.pdf https://www.danas.rs/bbc-news-serbian/trziste-kapitala-u-srbiji-sta-radi-i-cemu-sluzi-beogradska-berza/ https://www.danas.rs/bbc-news-serbian/trziste-kapitala-u-srbiji-sta-radi-i-cemu-sluzi-beogradska-berza/ https://www.belex.rs/ Bojan Đorđević, Sunčica Stanković 29 Bora, Debakshi, and Daisy Basistha. 2021. “The outbreak of COVID-19 pandemic and its impact on stock market volatility: Evidence from a worst-affected economy.” Journal of Public Affair, 1-10. https://doi.org/10.1002/pa.2623 Borjigin, Sumuya, Ting Gao, Yafei Sun, and Biao An. 2020. “For evil news rides fast, while good news baits later?—A network based analysis in Chinese stock market.” Physica A: Statistical Mechanics and its Applications, 551: 1-18. https://doi.org/10.1016/j.physa.2020.124593 Brueckner, Markus, and Joaquin Vespignani. 2021. “COVID-19 Infections and the Performance of the Stock Market: An Empirical Analysis for Australia. Economic Papers.” The Economic Society of Australia, 40(3): 173-193. https://doi.org/10.1111/1759-3441.12318 Claessens, Stijn, Hui Tong, and Shang-Jin Wei. 2012. “From the financial crisis to the real economy: Using firm-level data to identify transmission channels.” Journal of international economics, 88(2): 375–387. https://doi.org/10.1016/j.jinteco.2012.02.015 Dougherty, Christopher. 2011. Introduction to Econometrics. (4th edition). Oxford University Press. Đorđević, Bojan, and Sunčica Stanković. 2021. “Analiza uticaja pandemije COVID-19 na volatilnost srpskog tržišta kapitala.” Zbornik radova 11. Međunarodnog simpozijuma o upravljanju prirodnim resursima, Fakultet za menadžment, Zaječar, 90-97. UDK 616.98:578.834]:336.76(497.11) ISBN 978-86-7747-641-0 Field, Andy. 2005. Discovering statistics using SPSS. (4th Ed.). London: Sage Publications Ltd. Höhler, Julia, and Alfons O. Lansink. 2020. „Measuring the impact of COVID-19 on stock prices and profits in the food supply chain.” Agribusiness, 1-16. https://doi.org/10.1002/agr.21678 Hsua, Yu-Lin, and Li-Kai (Connie) Liao. 2022. “Corporate Governance and Stock Performance: The Case of COVID-19 Crisis.” Journal of Accounting and Public Policy, 41(4): 106920. https://doi.org/10.1016/j.jaccpubpol.2021.106920 Kenourgios, Dimitris, and Dimitrios Dimitrou. 2015. “Contagion of the Global Financial Crisis and the real economy: A regional analysis.” Economic Modelling, 44: 283–293. https://doi.org/10.1016/j.econmod.2014.10.048 Khalid, Noreen, Raja F. Zafar, Qasim R. Syed, and Roni Bhowmik. 2021. “The Heterogeneous Effects of COVID-19 Outbreak on Stock Market Returns and Volatility: Evidence from Panel Quantile Regression Model.” Etikonomi, 20(2): 225–238, https://doi.org/10.15408/etk.v20i2.20587 Lee, Kuo-Jung, and Su-Lien Lu. 2021. “The Impact of COVID-19 on the Stock Price of Socially Responsible Enterprises: An Empirical Study in Taiwan Stock Market.” International Journal of Environmental Research and Public Health, 18(4): 1398. https://doi.org/10.3390/ijerph18041398 MacKinlay, A. Craig. 1997. “Event studies in economics and finance.” Journal of Economic Literature, 35(1): 13–39. Naseem, Sobia, Muhammad Mohsin, Wang Hui, Geng Liyan, and Kun Penglai. 2021. “The Investor Psychology and Stock Market Behavior during the Initial Era of COVID-19: A Study of China, Japan, and the United States.” Frontiers in Psychology, 12:626934. https://doi.org/10.3389/fpsyg.2021.626934 Okorie, David I., Boqiang Lin. 2021. „Stock markets and the COVID-19 fractal contagion effects.” Finance Research Letters, 38: 101640. https://doi.org/10.1016/j.frl.2020.101640 Onali, Enrico. 2020. “Covid-19 and stock market volatility.” https://doi.org/10.2139/ssrn.3571453 Ozkan, Oktay. 2021. “Impact of COVID-19 on stock market efficiency: Evidence from developed countries.” Research in International Business and Finance, 58: 101445. https://doi.org/10.1016/j.ribaf.2021.101445 Özturk, Özcan, Muhammet Y. Şisman, Hakan Uslu, and Ferhat Çitak. 2020. “Effect of COVID- 19 outbreak on Turkish stock market: a sectoral-level analysis.” Hitit University Journal of Social Sciences Institute, 13(1): 56-68. https://doi.org/10.17218.hititsosbil.728146 https://doi.org/10.1002/pa.2623 https://doi.org/10.1016/j.physa.2020.124593 https://doi.org/10.1111/1759-3441.12318 https://doi.org/10.1016/j.jinteco.2012.02.015 https://doi.org/10.1002/agr.21678 https://doi.org/10.1016/j.jaccpubpol.2021.106920 https://doi.org/10.1016/j.econmod.2014.10.048 https://doi.org/10.15408/etk.v20i2.20587 https://doi.org/10.3390/ijerph18041398 https://doi.org/10.3389/fpsyg.2021.626934 https://doi.org/10.1016/j.frl.2020.101640 https://doi.org/10.2139/ssrn.3571453 https://doi.org/10.1016/j.ribaf.2021.101445 https://doi.org/10.17218.hititsosbil.728146 30 Economic Analysis (2022, Vol. 55, No. 2, 20-30) Park, Cyn-Young, and Kwanho Shin. 2020. “Contagion through National and Regional Exposures to Foreign Banks during the Global Financial Crisis.” Journal of Financial Stability, 46:100721. https://doi.org/10.1016/j.jfs.2019.100721 Ramelli, Stefano, and Alexander F. Wagner. 2020. “Feverish stock price reactions to covid- 19.” The Review of Corporate Finance Studies (online), 9(3): 622–655. https://doi.org/10.1093/rcfs/cfaa012 Randewich, Noel. 2020. “Wall Street dazed and confused after the worst day since 1987.” Reuters. https://www.reuters.com/article/us-health-coronavirus-markets-chaos- idUSKBN20Z3WB (accessed June 3, 2022). Sahoo, Manamani. 2021. “COVID-19 impact on stock market: Evidence from the Indian stock market.” Journal of Public Affair, 1–13. https://doi.org/10.1002/pa.2621 Satar, Muhammad A., Felix E. Arcilla Jr., and Muhammad F. Sattar. 2020. “The response of financial market indices to a covid-19 pandemic.” Financial Studies, 24(3): 83–92. Romanian Academy, National Institue of Economic Research (INCE) - Centre for Financial and Monetary Research. Singh, Bhanwar, Rosy Dhall, Sahil Narang, and Savita Rawat. 2020. “The Outbreak of COVID- 19 and Stock Market Responses: An Event Study and Panel Data Analysis for G-20 Countries.” Global Business Review, 1–26. https://doi.org/10.1177/0972150920957274 Tapaloglu, Emre E., Ilhan Ege, and Erol Koycu. 2021. „Coronavirus (Covid-19) and Stock Market: Empirical Analysis with Panel Data Approach.” International Journal of Economics and Finance, 13(3): 31–39. https://doi.org/10.5539/ijef.v13n3p31 Wang, Zijun, Victoria Salin, , Neal H. Hooker, and David Leathham. 2002. “Stock market reaction to food recalls: A GARCH application.” Applied economics letters, 9(15): 979–987. https://doi.org/10.1080/13504850210148125 West, Kenneth D. 1988. Dividend innovations and stock price volatility. Econometrica, 56(1): 37–61. https://doi.org/10.2307/1911841 Waheed, Rida, Suleman Sarwar, Sahar Sarwar, and Muhammad K. Khan. 2020. “The impact of COVID-19 on Karachi stock exchange: Quantile- on-quantile approach using secondary and predicted data.” Journal of Public Affairs, 20(4): e2290. https://doi.org/10.1002/pa.2290 World Health Organization. 2022. WHO Coronavirus (COVID-19) Dashboard [Dataset]. Retrieved from https://covid19.who.int/ Zaremba, Adam, Renatas Kizys, , David Y. Aharon, and Ender Demir. 2020. “Infected Markets: Novel Coronavirus, Government Interventions, and Stock Return Volatility around the Globe.” Finance Research Letters, 35: 101597. https://doi.org/10.1016/j.frl.2020.101597 Zeren, Feyyaz, and Atike Hizarci. 2020. “The impact of covid-19 coronavirus on stock markets: Evidence from selected countries.” Bulletin of Accounting and Finance Reviews, 3(1): 78–84. https://doi.org/10.32951/mufider.70615 Živković, Aleksandra. 2022. “Forecast of Belex15 and Belexline Movement Using ARIMA Model.“ Economic Analysis, 55(1): 90–104. DOI: 10.28934/ea.22.55.1.pp90-104 Article history: Received: July 27, 2022 Revised: October 19, 2022 Accepted: October 21, 2022 https://doi.org/10.1016/j.jfs.2019.100721 https://doi.org/10.1093/rcfs/cfaa012 https://www.reuters.com/article/us-health-coronavirus-markets-chaos-idUSKBN20Z3WB https://www.reuters.com/article/us-health-coronavirus-markets-chaos-idUSKBN20Z3WB https://doi.org/10.1002/pa.2621 https://doi.org/10.1177/0972150920957274 https://doi.org/10.5539/ijef.v13n3p31 https://doi.org/10.1080/13504850210148125 https://doi.org/10.2307/1911841 https://doi.org/10.1002/pa.2290 https://covid19.who.int/ https://doi.org/10.1016/j.frl.2020.101597 https://doi.org/10.32951/mufider.70615