Impact of the religious festivity on the Tunis Stock Exchange

The month of Ramadan is anticipated to influence the behavior of the stock market, where the environment in Ramadan is different from other months. During the Ramadan month, transformations in the social life of people are quite apparent and significant, and the overall economic activity tends to decelerate as the number of working hours decreases. This paper aims to explore the impact of the Ramadan month on the stock market returns, volatility, and trading volume on the Tunis Stock Exchange (TSE) between September 2009 and July 2019. To achieve these objectives, the Generalized Autoregressive Conditional Heteroskedasticity (GARCH) technique and the ordinary least squares regression are used. The results show that the impact of the Ramadan month on the daily returns is positive and statistically significant at a 1% level. It is also found that the impact of the Ramadan month on the volatility and trading volume is negative and statistically significant. The findings can help domestic and foreign investors and regulators to better comprehend the Tunisian stock exchange and investor behavior. Moreover, this research can help investors to develop their trading strategies.


INTRODUCTION
An eminent task in the financial literature has been to check whether seasonal effects are present in asset returns. A foreseeable effect is a contradiction to the market efficiency and can provide market participants with useful guidance on their investment decisions. Since the 1980s, numerous studies have recorded calendar effects in asset returns. Some of the most eminent seasonal effects include the month of the year effect and the day of the week effect (e.g., Rozeff & Kinney, 1976; Wang  Contrary to the January and Monday effects, which have been largely analyzed, the effect of Ramadan on equity markets remains generally neglected. Ramadan is the ninth month of the Hijri calendar. It moves ten or eleven days forward each year. The majority of Muslim countries utilize both the Hijri and the Gregorian calendars. The Gregorian calendar is utilized by businesses and governments. However, the Hijri calendar is utilized principally for the observance of religious activities and holidays. Throughout the Muslim states, during the month of Ramadan, the transformations in the social life of people are apparent and significant. Fasting Ramadan is one of the five pillars of Islam and is obligatory for every mature Muslim who is physically and mentally able to do it. During the month of Ramadan, all adult Muslims fast between sunrise and sunset, refraining from drink, food and other sensual pleasures and are heartened to devote themselves to spiritual training and almsgiving. Ramadan is the month of the Quran and returning to Allah with honest repentance so that it may remove our sins and immoral deeds.
Generally, during the Ramadan month, the economic activities are inclined to decelerate with a decrease in the number of working hours in the majority of sectors. However, the consumption of electricity tends to go up due to the augmentation of commercial and religious activities after the "Iftar". Moreover, the sales of groceries increase in Ramadan due to the evening "Iftar" feasts. Trading in stocks is recorded to reduce as many Muslims consider stock market speculation as a type of gambling, which is forbidden by Islam. Furthermore, in Ramadan, the employment of margin trading may reduce because it is heavily restricted under Sharia. In this context, the research problem in this paper can be summed up as follows: Does the month of Ramadan influence the behavior of the stock market, since the environment in Ramadan is not similar compared to other months? More specifically, does the month of Ramadan have a considerable impact on the stock market return, volatility and trading volume?

LITERATURE REVIEW AND HYPOTHESES
The Ramadan effect indicates that asset returns are considerably larger and less volatile on the month of Ramadan than the rest of the year in Islamic stock markets. Husain (1998)  To answer the research problem, three hypotheses will be tested: H1: The month of Ramadan has a significant impact on equity market returns.
H2: The month of Ramadan has a significant impact on the volatility of returns.
H3: The month of Ramadan has a significant impact on the trading volume.

DATA
The TSE is a small stock market in the African continent. Since the mid-1990s, it has become active as a result of regulatory reforms and privatization of public companies. It also shows a considerable non-normality for the Tunindex. The skewness coefficient is near to zero, but the kurtosis coefficient is excessively larger than three, signifying fat tails in comparison with a normal distribution. Moreover, the Jarque-Bera normality test statistics reject the normality of the daily returns of Tunindex at the one percent level. been widely employed to model financial time series and has confirmed to be prosperous in anticipating the conditional variances. The greatest frequent formulation of GARCH argues that the most prognosticator of the variance in the following time is a weighted mean of the long-run mean variance, the variance anticipated for this time, and the most current squared errors catching any recent information, with reducing weights accredited to the lag squared errors.

METHODOLOGY
To explore the Ramadan effect on daily returns and volatility for the Tunindex, the GARCH technique is used. This methodology is applied by where Ramadan D takes the value 1 for the Ramadan period, and 0 for the non-Ramadan period. A significant t α in equation (1) is taken as evidence of the Ramadan effect on the daily returns for the Tu n i ndex .
For the time-varying volatility of returns, it is modeled as a GARCH (p, q) specification to estimate the coefficients of the conditional variance equation (2). p and q are lagged orders of the GARCH model. The conditional variance is incorporated by GARCH as a linear function of past squared errors and lagged conditional variances. To examine the impact of Ramadan on returns and volatility for Tunindex, the return (1) and volatility (2) equations are jointly estimated by applying the Full Information Maximum Likelihood method (FIML).

RESULTS AND DISCUSSION
The estimation findings of the GARCH (1, 1) technique for equations (1) and (2) for the Tunindex are documented separately in Table 2. Columns Columns (c) through (f) in Table 2 provide the estimation results of the volatility equation. The dummy parameter is negative and statistically significant at the 5% level, indicating that during the month of Ramadan, return volatility reduces significantly. The results confirm the second hypothesis (H2) tested in this research. The second hypothesis argues that the month of Ramadan has a significant impact on the volatility of the returns. The considerable decline in the volatility of returns during this month can be the outcome of diminished trading hours at the exchange and the investors' religious perception. In other words, many Muslim investors think that speculation in stock markets is a type of gambling, which is unacceptable in Islam. These findings

CONCLUSION
The study of the Ramadan effect on the stock market allows investors to understand how they design their trading strategies. This paper explores the Ramadan effect on the Tunis Stock Exchange (TSE) between September 2009 and July 2019. It documents evidence of the Ramadan effect on the daily returns, volatility and trading volume by applying the GARCH model and the ordinary least squares regression. More specifically, the paper reports a considerable increase in the returns during the month of Ramadan. Also, it is found that return volatility and trading volume reduce significantly during this month. The significant increase in returns in the month of Ramadan may be justified by the fact that Tunisian investors are optimistic about the month of Ramadan, making them more likely to purchase stocks than to sell them. The existence of the Ramadan anomaly in Tunindex daily returns may challenge the efficient market hypothesis and supports the behavioral finance literature because it allows investors to achieve abnormal returns by buying stocks on non-Ramadan days and selling them during the month of Ramadan. Moreover, a considerable decrease in the volatility of returns and the trading volume in the month of Ramadan can be attributed to the diminution in the number of trading hours on the TSE and the investors' religious perception. More specifically, many Muslim investors think that speculation in stock markets is a type of gambling, which is unacceptable in Islam. The presence of a considerable decline in the volatility of returns during the month of Ramadan indicates foreseeable changes in the price of risk. The existence of the Ramadan effect should not automatically imply that excess profits can be generated on the TSE, because transaction costs can render any investment strategy dependent on the Ramadan effect non-lucrative. The results found in this paper can help domestic and foreign investors and regulators to better understand the Tunisian stock market and investor behavior. Moreover, this research can help investors develop their trading strategies. An important extension of this paper would be an analysis of the existence of the Ramadan effect on the TSE using sector indices.