Hijri calendar effect in Borsa Istanbul gold market and Turkey’s foreign exchange market

Nasif Ozkan (Department of Banking and Finance, Kutahya Dumlupinar University, Kutahya, Turkey)

Journal of Islamic Accounting and Business Research

ISSN: 1759-0817

Article publication date: 8 July 2019

Abstract

Purpose

This study aims to investigate the Hijri calendar effect in Borsa Istanbul (BIST) precious metal market and foreign exchange market (Dollar and Euro market) of Turkey.

Design/methodology/approach

The data of BIST gold market index and foreign exchange market are used for the period of 4 March 2003-30 September 2016 (1 Muharram 1424 – 28 Dhu al-Hijja 1437) in the study. These data are analyzed by using the dummy variable regression model and Kruskal–Wallis (KW) test.

Findings

The results of the regression models and KW test indicate that there is a Ramadan effect in the gold market and after-Ramadan effect in the Euro market. On the other hand, the Hijri month effect does not exist in the Dollar market.

Originality/value

This is the first paper that investigates the Hijri calendar effect in gold and foreign exchange markets of Turkey other than the stock market.

Keywords

Citation

Ozkan, N. (2019), "Hijri calendar effect in Borsa Istanbul gold market and Turkey’s foreign exchange market", Journal of Islamic Accounting and Business Research, Vol. 10 No. 4, pp. 580-590. https://doi.org/10.1108/JIABR-04-2017-0054

Publisher

:

Emerald Publishing Limited

Copyright © 2019, Emerald Publishing Limited


1. Introduction

After the emergence of the efficient market hypothesis, many researchers investigated whether stock markets were efficient or not and they provided evidence that stock markets were not efficient, which means that stock prices did not fully reflect all the information and expectations about the stocks. Therefore, investors could apply various trading strategies in the stock market to obtain abnormal returns. All these findings, which are contrary to the efficient market hypothesis, are defined as “anomaly” in the literature. One of the most investigated anomalies in the stock markets is the calendar anomalies. Calendar anomalies indicate that stock returns are higher or lower on certain days, weeks and months. Within the calendar anomalies, the “January effect” has been the subject of many studies. These studies suggest that January provides higher returns than the other months of the year (Aytekin and Sakarya, 2014; Karan and Uygur, 2001; Kato and Schallheim, 1985; Rozeff and Kinney, 1976; Thaler, 1987).

In recent years, studies suggest that stock markets provide higher returns on some days, weeks or months determined by the Hijri calendar (day of Ashura, the month of Ramadan, etc.) than others (Białkowski et al., 2012; Majeed et al., 2015; Oğuzsoy and Güven, 2004). In literature, it is emphasized that the fasting month of Ramadan has significant effects on the economic and financial variables. During Ramadan, it is observed that the trading activities in the financial markets of the Muslim countries change noticeably and market participants rather fulfill their religious duties (Seyyed et al., 2005). Although the Gregorian calendar is used in many Muslim countries including in Turkey, the religious days are determined by the Hijri calendar. These two calendar systems are different from each other because of their number of days per year[1]. Therefore, the religious days and festivals in Muslim countries usually begin approximately 11 days earlier compared to each year of the Gregorian calendar. For all these reasons, the investigation of the Hijri calendar effect has great importance for the financial market regulators and participants. The existing studies in Turkey on this issue have been mostly carried out on Borsa Istanbul (BIST) (Akbalık, 2015; Küçüksille and Özmutaf, 2015; Oğuzsoy and Güven, 2004). Therefore, the aim of this study is to investigate the Hijri calendar effect in gold and foreign exchange market (both Dollar and Euro market). To investigate the presence of the Hijri calendar effect in these markets, the data from the BIST gold market index and Central Bank of Turkey (CBT) indicative selling prices of Dollar and Euro, are used for the period of 4 March 2003-30 September 2016 (1 Muharram 1424 – 28 Dhu al-Hijja 1437). These data are analyzed by using the dummy variable regression model and Kruskal–Wallis test (KW).

This paper is organized as follows: Section 2 reviews literature regarding the impact of the Islamic calendar events on financial markets; Section 3 explains the data; Section 4 describes the research methodology used in the study; Section 5 reports the empirical results; and Section 6 presents the results on robustness checks and Section 6 summarizes the findings.

2. Literature review

The month of the year effect’ is one of the calendar anomalies extensively examined in many studies in the finance literature. These studies reveal that stock returns are higher in some months of the year than others. January is the month that stocks have higher returns and this anomaly is called the “January effect” (Aytekin and Sakarya, 2014; Karan and Uygur, 2001; Kato and Schallheim, 1985; Rozeff and Kinney, 1976; Thaler, 1987). These studies used the Gregorian calendar to investigate the January effect in the financial markets. In recent years, there have also been studies investigating whether stocks provide significantly different average returns in Hijri months of the year based on the Hijri calendar. In most of these studies, it is ascertained that stocks provide higher returns in the fasting month of Ramadan and this is called “Ramadan effect” in the finance literature. Table I provides a brief overview of relevant studies focusing on the Ramadan effect and the other Hijri calendar effects on stock markets.

The first study investigating the Ramadan effect is Husain (1998). Husain (1998) examines the Ramadan effect on Pakistan’s stock market between the years 1989 to 1993 using dummy variable regression and generalized autoregressive conditional heteroskedasticity (GARCH) models. The author suggests that there is a significant decline in stock returns volatility in Ramadan while the mean return remains the same. In another study, Seyyed et al. (2005) use the GARCH model to investigate the “Ramadan effect” in the Saudi Arabian stock market and reach similar findings to those of Husain (1998). Ariss et al. (2011) are also reported similar findings for the Arab Gulf Cooperation Council countries’ stock markets. On the other hand, Ramezani et al. (2013) demonstrate that there is a statistically significant positive (negative) relationship between the changes of the Tehran stock market index and Ramadan, Shawwal and Rabi I (Jumada II, Rabi II, Muharram and Rajab) months using the GARCH model. The authors also reveal that there is no significant relationship between the changes of the Tehran stock market index and Jumada I, Safer, Sha’ban, Dhu al-Qa’da and Dhu al-Hijja months.

Białkowski et al. (2012) investigate the presence of the Ramadan effect in the stock markets of 14 Muslim countries for the period from 1989 to 2007. The authors assert that stock returns are almost nine times higher and less variable in Ramadan compared to the other months of the Hijri calendar. Al-Hajieh et al. (2011) prove the presence of the Ramadan effect in many Muslim countries stock market located in the Middle East. The authors also state that market returns in the first and the last days of Ramadan are highly statistically significant year on year variation. Al-Hajieh et al. (2011) suggest that this can be attributed to investor sentiment or positive investor mood.

Almudhaf (2012) examines the significance of the seasonal impact of the Hijri calendar between the years 1996-2007 in 12 Muslim countries. The author emphasizes that investors achieve positive and statistically significant returns in Kuwait, Jordan, Pakistan and Turkey stock markets in Ramadan. Mustafa (2011) investigates the Hijri month effect in the Pakistan stock market for the period of December 1991- 2010 by using the ordinary least square method (OLS). Parallel with the findings of Almudhaf (2012), the author reveals the existence of the “Ramadan effect” on the Pakistan stock market. In addition, Iqbal et al. (2013) examine conventional and Islamic calendar anomalies in the Pakistan stock market by using OLS method and indicate the existence of the “day of the week effect”, “month of the year effect”, “end of the month effect”, “half month effect”, “Islamic month effect” and the “Ramadan effect” between the years 1992 and 2011. In contrast to the findings of Almudhaf (2012) and Küçüksille and Özmutaf (2015) ascertain that there is not a “Ramadan effect” in BIST during the period of 1988-2014. On the other hand, the authors demonstrate that the month of Rajab has the highest, the month of Rabi II has the lowest and the month of Ramadan has the fourth highest mean returns during the examined period. Akbalık (2015), who investigates the existence of the Ramadan effect in all BIST indices, exhibits significant Ramadan effect only for the BIST services index and BIST transportation index. The author also suggests that the “Ramadan effect” has been diminishing in recent years.

Białkowski et al. (2013) examine whether the mutual fund managers investing in Turkish stocks benefit from the Ramadan effect. The authors show that domestic institutional funds, hybrid funds and foreign Turkish equity funds have higher risk-adjusted performances during the month of Ramadan compared to other months of the Hijri year. On the other hand, Białkowski et al. (2013) state that domestic index funds are adversely affected by the month of Ramadan due to the increased money inflows in Ramadan and could not provide higher abnormal returns. Alrashidi et al. (2014) examine the existence of the Ramadan effect on the 52 Islamic equity funds within the period of 2004-2009. The authors indicate that although there is not a significant change in the returns of Islamic equity funds in Ramadan, the return volatility declines.

Oğuzsoy and Güven (2004) study the effect of the religious holidays (the feast of Ramadan and the feast of Sacrifice) on stock returns in BIST for the period between 1988 and 1999. The authors reveal that the BIST 100 index provide higher returns two days before religious holidays. They also indicate that 17 of the BIST 30 index stocks display seven times higher average returns on two days before religious holidays compared to the average returns on other days. In addition, Oğuzsoy and Güven (2004) demonstrate that the daily return performance for 21 of the BIST 30 index stocks is more than 2 per cent on holy days. Majeed et al. (2015) investigate the effect of Islamic calendar events (day of Ashura, the feast of Ramadan, the feast of Sacrifice and the birthday of the Prophet Muhammed) on Pakistan stock market and they specify that all events except the feast of Sacrifice have a statistically significant effect on stock returns. The authors claim that investors are able to gain statistically significant abnormal returns before the month of Ramadan, day of Ashura, the birthday of the Prophet Muhammed and the feast of Ramadan. On the other hand, statistically significant abnormal returns are observed only after the month of Ramadan and the feast of Ramadan. Majeed et al. (2015) also indicate that the month of Muharram and the month of Ramadan have a significant positive effect on Pakistan’s stock market. Al-Ississ (2010) examines the effect of Muslim holy days (the month of Ramadan and the day of Ashura) on the daily returns and trading volumes of 17 Muslim financial markets for the period between 1988 and 2009. The author states that trading volumes of financial markets drop statistically significantly in Muslim holy days. Al-Ississ (2010) also indicates that the daily returns of financial markets are affected positively by the month of Ramadan and negatively by the day of Ashura.

3. Data

In this study, the data of BIST precious metals market and foreign exchange market are used for the period of 4 March 2003-30 September 2016 (1 Muharram 1424 – 28 Dhu al-Hijja 1437). While BIST precious metals market data consist of daily closing prices of the BIST gold market index, the foreign exchange market data consists of daily indicative selling prices of Dollar and Euro determined by CBT the previous working day. All the data used in this study were obtained from the CBT electronic data delivery system. Daily returns related to BIST gold market index, Dollar and Euro are calculated with the help of the formula below:

(1) Rit=lnPitPit-1

In equation (1), Rit denotes the daily logarithmic rate of return for the investment instrument i on day t; Pit and Pit-1 represent the daily closing price of the investment instrument i on day t and t − 1, respectively and ln is the natural log. The reason for calculating the returns by taking their natural logarithm (In) is, to eliminate the negative effects caused by extreme values in the data set on the results (Ergül et al., 2009; Ergül et al., 2008; Tunçel, 2007). To avoid inappropriate interpretation of the analysis, all investments’ instruments returns are tested for stationary using the augmented Dickey–Fuller unit root test (ADF) (Dickey and Fuller, 1981).

4. Methodology

In the study, dummy variable regression techniques are used to test the effects of the Hijri month of the year. In this model, a dummy variable representing a specific month of the Hijri year and trading days of this specific month takes the value of 1 while the rest of the trading days in other Hijri months take the value of 0[2]. To test the Hijri month effect, the following regression model is estimated:

(2) Rit=β1iD1i+β2iD2i+β3iD3i++ β9iD9i+β10iD10i+β11iD11i+β12iD12i+εti

In equation (2), Rit represents the daily rate of return for the investment instrument i, D1i, …, D12i denote the dummy variables of investment instrument i for each Hijri month. For any investment instrument i, D1i takes the value of 1 if the Hijri month is Muharram, and 0 otherwise; D2i takes the value of 1 if the Hijri month is Safar, and 0 otherwise; ….; D12i takes the value of 1 if the Hijri month is Dhu al-Hijja, and 0 otherwise. Betas (β1i,…, β12i) are the regression coefficients and εti is the error term of the regression model. If the β9i coefficient is statistically significantly positive or statistically significantly bigger than the coefficients of other Hijri months, the presence of the Ramadan effect is proved for any investment instrument i.

5. Empirical results

Descriptive statistics of daily returns for the gold market index and currencies (Dollar and Euro) are presented in Table II. The average daily return of Gold, Dollar and Euro markets is not statistically significantly different from zero. In the analyzed period, the highest (43.4517 per cent) and the lowest return (−41.1601 per cent) is observed in the Gold market. The results of the normality test (Jarque–Bera statistics) and ADF for the return series of all variables are also shown in Table II. Jarque–Bera statistics for all variable returns demonstrate that the rejection of the normality assumption at 1 per cent significance level. These results refer that none of the time-series for the variables are normally distributed. In the ADF test, the null hypothesis (H0) that claims return series have unit roots is rejected at 1 per cent significance level for all variables. This means that return series are stationary at level.

The results of equation (2) that investigates the presence of the Hijri month effect in the Gold and foreign exchange markets are shown in Table III. If the coefficients of the dummy variables for the Gold market are individually examined, it can be seen that Gold market has statistically significant positive returns in the month of Rabi I at the 5 per cent level and Ramadan at the 1 per cent level (β3Gold = 0.1422 per cent and β9Gold = 0.1842 per cent). In parallel with the literature (Al-Hajieh et al., 2011; Almudhaf, 2012; Białkowski et al., 2012; Mustafa, 2011), Gold market provides higher returns in Ramadan compared to other months of the Hijri year. On the other hand, the findings on the Dollar market indicate that this currency provides statistically significant positive return in the month of Shàban and Shawwal at the 10 per cent level (β8Dollar = 0.0720 per cent and β10Dollar = 0.1715 per cent). The findings for the Euro market show that this currency provides a statistically significant positive return in the month of Rabi I similar to Gold market (β3Euro = 0.0922 per cent) and provides a statistically significant positive return in the month of Shawwal similar to the Dollar market (β10Euro = 0.1832 per cent). As a result, both currencies’ coefficients of dummy variables for the month of Shawwal are statistically significantly positive. Therefore, these findings promise that there is an after-Ramadan effect in the foreign exchange market of Turkey.

6. Robustness test

The dummy variable regression model is implemented by assuming that the dependent variables of the regressions are normally distributed. However, the return series may not follow a normally distributed pattern. The Jarque–Bera statistics results shown in Table II reveal that the return series clearly violate the normality assumption. Therefore, a non-parametric test, KW, is used to verify the findings of the dummy variable regression. KW test is known as a non-parametric alternative of one-way analysis of variance (ANOVA). This test provides a comparison of three or more groups with continuous variables (Kalaycı, 2010). In the KW test, KW value chi-square (χ2) statistic is calculated by using the following formula (Hui, 2005; Kumar and Pathak, 2016; Lim and Chia, 2010):

(3) KW=12n(n+1)j=1kRj2nj-3(n+1)

In equation (3), Rj is the rank sum of observations in the jth group, n is the total number of sample observations, nj is the total number of observations in the jth group and k is the number of groups. Similar to Hui (2005) and Kumar and Pathak (2016), the following hypothesis are generated for testing the presence of the Hijri month effect in each investment instrument (Gold, Dollar and Euro) using KW test:

H0.

There is no difference between the mean returns of each investment instruments (Gold, Dollar and Euro) in Hijri months.

H1.

There is a difference between the mean returns of each investment instruments (Gold, Dollar and Euro) in Hijri months.

As a result of the KW test, rejection of theH0, for an investment instrument means that there is a Hijri month effect in that market. In other words, such a result implies that there is a difference between the mean returns of the investment instrument in Hijri months.

Table IV shows the results of the KW test for all investment instruments. The results for Gold market and Euro market are similar to the results obtained from the dummy variable regression model presented in Table III. χ2 statistics reveal the presence of the Hijri month effect in the Gold market and the Euro market at 10 per cent and 1 per cent level, respectively (χ2Gold = 19.2860, p-valueGold = 0.0560; χ2Euro = 25.5620, p-valueEuro = 0.0080). On the other hand, there is insufficient evidence to prove the presence of the Hijri month effect in the Dollar market. All of these results support the presence of the Ramadan effect in the Gold market and after-Ramadan effect in the Euro market.

7. Conclusion

The Hijri calendar effect on the returns of the investment instruments is investigated in many studies conducted in Muslim countries. In reviewing the literature on this subject, the studies are more focused on stock markets. These studies assert that stocks provide the highest return during the ninth month of the Hijri year called Ramadan. In this study, the data of BIST gold market index and foreign exchange market is used for the period of 4 March 2003-30 September 2016 (1 Muharram 1424 – 28 Dhu al-Hijja 1437) to investigate the presence of the Hijri calendar effect in Turkey. Therefore, the study differs from the other Hijri calendar effect studies in terms of data examined. To find out the Hijri calendar effect, the dummy variable regression model and KW are used. The results of the regression models and the KW test indicate that there is a Ramadan effect in the Gold market and after-Ramadan effect in the Euro market. Specifically, Gold market investors gain higher returns over the fasting month of Ramadan than the rest of the Hijri year in line with the stock market findings of those Al-Hajieh et al. (2011), Białkowski et al. (2012) and Mustafa (2011). These results also indicate that Euro market investors may realize the special pattern in Euro market behavior over the month of Shawwal. On the other hand, the presence of the Hijri month effect cannot be proven in the Dollar market. The above-mentioned findings may serve as a guide for investors to buy and sell Gold and Euro. In these cases, if we do not consider the transaction costs, employing a buy and hold strategy of buying on the last day of the month before Ramadan (Shawwal) and selling on the last trading day of Ramadan (Shawwal) in the Gold (Euro) market an investor may make positive returns over time.

While Dollar and Euro, the base currencies traded in Turkish financial markets, are analyzed, other currencies (such as British Pound, Japanese Yen, etc.) have not been included in this study. In addition, there are various econometric methods (i.e. GARCH models) other than dummy variable regression used by the researchers (Kiymaz and Berument, 2003; Seyyed et al., 2005) to test the existence of the calendar anomalies in stock markets. Thus, future studies may cover all currencies traded in Turkish financial markets and apply to other methods to test the existence of the Hijri calendar effect in those markets. Therefore, the foregoing study will constitute an important reference point for future studies.

The Studies on Ramadan effect and the other Hijri calendar effects

Author (year) Country/data set Study period Method Findings
Husain (1998) Pakistan (36 individual stocks, 8 sector indices and the KSE 100 index) 1989-1993 Dummy variable regression and GARCH models While the Ramadan has no significant effect on the stock returns, the volatility of stock returns decline significantly in this month
Oğuzsoy and Güven (2004) Turkey (BIST 100 index and BIST 30 index stocks) 1988-1999 Dummy variable regression and statistical tests BIST 100 index and 17 of the BIST 30 index stocks provide higher returns two days before religious holidays. In addition, 21 of the BIST 30 index stocks’ daily return performance is more than 2% on holy days
Seyyed et al. (2005) Saudi Arabia (the main stock market index and sector indices such as banking, industry, cement, electricity, agriculture and services) 1985-2000 GARCH model While the Ramadan has no significant effect on the market returns, the volatility of market returns decline significantly in this month
Al-Ississ (2010) The main stock market indices of 17 Muslim countries such as Bahrain, Egypt, Indonesia, Iran, Jordan, Kuwait, Lebanon, Malaysia, Morocco, Oman, Pakistan, Palestine, Qatar, Saudi Arabia, Tunisia, Turkey and the United Arab Emirates 1988-2009 Panel data regression Trading volumes of 17 Muslim stock markets decline significantly in the month of Ramadan and the day of Ashura. In addition, Ramadan (the day of Ashura) has positive (negative) effect on the daily returns of 17 Muslim stock markets
Al-Hajieh et al. (2011) The main stock market indices of Bahrain, Egypt, Jordan, Kuwait, Qatar, Saudi Arabia, Turkey and the United Arab Emirates 1992-2007 Runs test There is a Ramadan effect in the stock market of those countries and market returns in the first and the last days of Ramadan are statistically significant year on year variation
Ariss et al. (2011) The main stock market indices of GCC countries such as Bahrain, Kuwait, Oman, Qatar, Saudi Arabia and the United Arab Emirates (Abu Dhabi and Dubai) 1994-2008 Dummy variable regression While the Ramadan has no significant effect on the market returns, the volatility of market returns decline significantly in this month
Mustafa (2011) Pakistan (KSE 100 index) 1991-2010 Dummy variable regression Ramadan has a significant effect on stock market returns and increases the stock market risk
Almudhaf (2012) The main stock market indices of 12 Muslim countries such as Bahrain, Egypt, Indonesia, Jordan, Kuwait, Malaysia, Morocco, Oman, Pakistan, Saudi Arabia, Turkey and the United Arab Emirates 1996-2007 Dummy variable regression Ramadan has a significant effect on stock market returns of Kuwait, Jordan, Pakistan and Turkey
Białkowski et al. (2012) The main stock market indices of 14 Muslim countries such as Bahrain, Egypt, Indonesia, Jordan, Kuwait, Malaysia, Morocco, Oman, Pakistan, Qatar, Saudi Arabia, Tunisia, Turkey and the United Arab Emirates 1989-2007 Statistical tests and panel data regression Ramadan has a strong effect on the stock market returns
Białkowski et al. (2013) Turkey (Turkish mutual funds and foreign mutual funds with an investment focus on Turkish equities) 2000-2011 GARCH models, performance measurement models and total performance models Ramadan has a significant positive effect on the risk-adjusted performance of domestic institutional funds, hybrid funds and foreign Turkish equity funds
Iqbal et al. (2013) Pakistan (KSE 100 index) 1992-2011 Dummy variable regression Ramadan has a significant effect on stock market returns and the volatility of market returns decline significantly in this month
Ramezani et al. (2013) Iran (Tehran Stock Exchange) 2002-2012 Autoregressive conditional heteroscedasticity, GARCH, autoregressive moving average and autoregressive integrated moving average models While Ramadan, Shawwal and Rabi I has significant positive effect on the market returns; Jumada II, Rabi II, Muharram and Rajab has significant negative effect on it
Alrashidi et al. (2014) Globally selected 52 Islamic mutual funds 2004-2009 Dummy variable regression While the Ramadan has no significant effect on the performance of Islamic equity funds, the volatility of stock returns decline significantly in this month
Akbalık (2015) Turkey (BIST 100, BIST 30, sector and sub-sector indices) 1997-2013 Statistical tests Ramadan only has a significant effect on the BIST services index and BIST transportation index
Küçüksille and Özmutaf (2015) Turkey (BIST 100 index) 1988-2014 Statistical tests Ramadan provides the fourth highest mean return in the market
Majeed et al. (2015) Pakistan (KSE 100 index) 2001-2012 Dummy variable regression and event study The day of Ashura, the feast of Ramadan and the birthday of Prophet Muhammed have significant effect on stock returns

Descriptive statistics of daily percentage returns for Gold, Dollar and Euro markets

Descriptive Gold Dollar Euro
Mean 0.0395 0.0185 0.0196
Maximum 43.4517 7.0429 4.7709
Minimum −41.1601 −11.9348 −67.718
Standard deviation 0.0157 0.0083 0.0081
Skewness 0.8896 −0.0530 0.3680
Kurtosis 319.5133 20.4138 8.8001
Jarque–Bera 14,113,398*** 43,200.81*** 4,869.663***
ADF −48.3600*** −57.3311*** −56.8834***
Observations 3,381 3,419 3,419
Note:

*** Represents the rejection of null hypothesis at 1% significance level

Regression results of Hijri month effect for Gold, Dollar and Euro markets

Month Gold Dollar Euro
Muharram −0.0325 (−0.4676) 0.0430 (0.9407) −0.0113 (−0.2917)
Safar −0.0012 (−0.0226) 0.0614 (1.3820) 0.0748 (1.5364)
Rabi I 0.1422** (2.3416) 0.0407 (0.7659) 0.0922* (1.8952)
Rabi II −0.0500 (−0.7013) 0.0272 (0.4498) −0.0108 (−0.1655)
Jumada I 0.0610 (0.9665) −0.0172 (−0.3931) −0.0428 (−1.0517)
Jumada II 0.0057 (0.0857) −0.0021 (−0.0451) −0.0126 (−0.3271)
Rajab 0.0114 (0.2322) −0.0457 (−1.0683) 0.0166 (0.3976)
Sha’ban −0.0397 (−0.5706) 0.0720* (1.8270) 0.0348 (0.7850)
Ramadan 0.1842*** (2.6688) −0.0495 (−1.2466) −0.0290 (−0.7219)
Shawwal 0.1004 (1.0267) 0.1715* (1.9424) 0.1832*** (3.4387)
Dhu al-Qa’da 0.0129 (0.1391) −0.0655 (−1.0324) −0.0125 (−0.2347)
Dhu al-Hijja 0.1007 (1.4917) −0.0017 (−0.0409) −0.0388 (−0.9109)
R2 0.0021 0.0055 0.0060
DW 2.4461 1.9621 1.9451
Notes:

The table shows the estimation results for equation (2) of daily percentage returns for Gold, Dollar and Euro. The regressions are estimated using OLS adjusted for Newey and West (1987) heteroscedasticity-consistent covariance matrix. Figures in the parenthesis show the t-statistics.

***;

**

; and

* represents statistical significance at 1, 5 and 10% levels, respectively

KW test results for Gold, Dollar and Euro markets

Asset χ2 statistics p-value H0
Gold 19.2860* 0.0560 Reject
Dollar 13.2810 0.2750 Accept
Euro 25.5620*** 0.0080 Reject
Notes:

The H0 states that there is no difference in the returns across the months of the Hijri years. If the H0 is rejected, this implies that there is a Hijri month effect in the investment instrument.

***; and

* represent statistical significance at 1 and 10% level

Notes

1.

Hijri calendar is a lunar calendar and begins with the emigration of Prophet Muhammed (pbuh) from Mecca to Medina, known as the Hijra. One year consists of 354 days and 12 months in Hijri calendar system. The Gregorian calendar currently used in Turkey is the solar calendar and begins with the birth of Prophet Jesus (pbuh). In this calendar system, the year is based on the revolution of the Earth around the Sun and it takes 365 days and 6 hours (Unat, 2004).

2.

Hijri calendar is a lunar calendar consisting of 12 months in a year of 354 days. The months of the Hijri calendar can be listed as Muharram, Safar, Rabi I, Rabi II, Jumada I, Jumada II, Rajab, Sha’ban, Ramadan, Shawwal, Dhu al-Qa’da and Dhu al-Hijja. Ramadan is the ninth month of the Hijri calendar.

References

Akbalık, M. (2015), “An analysis of the effect of Ramadan on Istanbul stock exchange”, in Yılmaz, R., Löschnigg, G., Arslan, H. and Icbay, M.A. (Eds), Current Approaches in Social Sciences, Peter Lang GmbH, Peter Lang, pp. 573-579.

Al-Hajieh, H., Redhead, K. and Rodgers, T. (2011), “Investor sentiment and calendar anomaly effects: a case study of the impact of Ramadan on Islamic Middle Eastern markets”, Research in International Business and Finance, Vol. 25 No. 3, pp. 345-356.

Al-Ississ, M. (2010), The Impact of Religious Experience on Financial Markets, Harvard Kennedy School of Government, Cambridge, MA.

Almudhaf, F. (2012), “The Islamic calendar effects: evidence from twelve stock markets”, International Research Journal of Finance and Economics, No. 87, pp. 185-191.

Alrashidi, F., Ahmed, M. and Beneid, F. (2014), “The calendar impact and trading behavior: an empirical evidence from around the globe”, International Business and Economics Research Journal (IBER), Vol. 13 No. 5, pp. 1025-1032.

Ariss, R.T., Rezvanian, R. and Mehdian, S.M. (2011), “Calendar anomalies in the Gulf cooperation council stock markets”, Emerging Markets Review, Vol. 12 No. 3, pp. 293-307.

Aytekin, S. and Sakarya, Ş. (2014), “Ocak ayı anomalisi: Borsa İstanbul endeksleri üzerine bir uygulama”, International Journal of Management Economics and Business, Vol. 10 No. 23, pp. 137-156.

Białkowski, J., Etebari, A. and Wisniewski, T.P. (2012), “Fast profits: investor sentiment and stock returns during Ramadan”, Journal of Banking and Finance, Vol. 36 No. 3, pp. 835-845.

Białkowski, J., Bohl, M.T., Kaufmann, P. and Wisniewski, T.P. (2013), “Do mutual fund managers exploit the Ramadan anomaly? Evidence from Turkey”, Emerging Markets Review, Vol. 15, pp. 211-232.

Dickey, D.A. and Fuller, W.A. (1981), “Likelihood ratio statistics for autoregressive time series with a unit root”, Econometrica, Vol. 49 No. 4, pp. 1057-1072.

Ergül, N., Akel, V. and Dumanoğlu, S. (2009), “Sektör endekslerinde haftanını günü etkisinin araştırılması”, Muhasebe Bilim Dünyası Dergisi, Vol. 11 No. 2, pp. 129-152.

Ergül, N., Dumanoğlu, S. and Akel, V. (2008), “İMKB’de günlük anomaliler”, Marmara Üniversitesi İ.İ.B.F. Dergisi, Vol. 25 No. 2, pp. 601-629.

Hui, T.-K. (2005), “Day-of-the-Week effects in US and Asia–Pacific stock markets during the Asian financial crisis: a non-parametric approach”, Omega, Vol. 33 No. 3, pp. 277-282.

Husain, F. (1998), “A seasonality in the Pakistani equity market: the Ramadhan effect”, The Pakistan Development Review, Vol. 37 No. 1, pp. 77-81.

Iqbal, M.S., Kouser, R. and Azeem, M. (2013), “Conventional and Islamic anomalies in Karachi stock exchange”, Science International, Vol. 25 No. 4, pp. 999-1007.

Kalaycı, Ş. (2010), SPSS Uygulamalı Çok Değişkenli İstatistik Teknikleri, Asil Yayın Dağıtım, Ankara.

Karan, M.B. and Uygur, A. (2001), “İstanbul menkul kıymetler borsası’nda haftanın günleri ve ocak ayı etkilerinin firma büyüklüğü açısından değerlendirilmesi”, Ankara Üniversitesi SBF Dergisi, Vol. 56 No. 2, pp. 103-115.

Kato, K. and Schallheim, J.S. (1985), “Seasonal and size anomalies in the Japanese stock market”, The Journal of Financial and Quantitative Analysis, Vol. 20 No. 2, pp. 243-260.

Kiymaz, H. and Berument, H. (2003), “The day of the week effect on stock market volatility and volume: international evidence”, Review of Financial Economics, Vol. 12 No. 4, pp. 363-380.

Küçüksille, E. and Özmutaf, N.M. (2015), “Is there Ramadan effect in Turkish stock market”, Uluslararası Alanya İşletme Fakültesi Dergisi, Vol. 7 No. 3, pp. 137-142.

Kumar, S. and Pathak, R. (2016), “Do the calendar anomalies still exist? Evidence from Indian currency market”, Managerial Finance, Vol. 42 No. 2, pp. 136-150.

Lim, S.Y. and Chia, R.C.-J. (2010), “Stock market calendar anomalies: evidence from Asean – 5 stock markets”, Economics Bulletin, Vol. 30 No. 2, pp. 996-1005.

Majeed, U., Raheman, A., Sohail, M.K., Bhatti, G.A. and Zulfiqar, B. (2015), “Islamic calender events and stock market reaction: evidence from Pakistan”, Science International, Vol. 27 No. 3, pp. 2559-2567.

Mustafa, K. (2011), “The Islamic calendar effect on Karachi stock market”, Pakistani Business Review, Vol. 13 No. 3, pp. 562-574.

Oğuzsoy, C.B. and Güven, S. (2004), “Holy days effect on Istanbul stock exchange”, Journal of Emerging Market Finance, Vol. 3 No. 1, pp. 63-75.

Ramezani, A., Pouraghajan, A. and Mardani, H. (2013), “Studying impact of Ramadan on stock exchange index: case of Iran”, World of Science Journal, Vol. 1 No. 12, pp. 46-54.

Rozeff, M.S. and Kinney Jr, W.R. (1976), “Capital market seasonality: the case of stock returns”, Journal of Financial Economics, Vol. 3 No. 44, pp. 379-402.

Seyyed, F.J., Abraham, A. and Al-Hajji, M. (2005), “Seasonality in stock returns and volatility: the Ramadan effect”, Research in International Business and Finance, Vol. 19 No. 3, pp. 374-383.

Thaler, R.H. (1987), “Anomalies: the January effect”, The Journal of Economic Perspectives, Vol. 1 No. 1, pp. 197-201.

Tunçel, A.K. (2007), “İMKB’de haftanın günü etkisi”, Akdeniz Üniversitesi ÍÍBF Dergisi, Vol. 13, pp. 252-265.

Unat, Y. (2004), “İslâm’da ve türklerde zaman ve takvim”, in Oğuz, Ö. (Ed.) Türk Dünyası, Nevruz Ansiklopedisi, Atatürk Kültür Merkezi Başkanlığı Yayınları, Ankara, pp. 15-24.

Corresponding author

Nasif Ozkan can be contacted at: nasif.ozkan@dpu.edu.tr

About the author

Nasif Ozkan is currently an Associate Professor at Kutahya Dumlupinar University, School of Applied Sciences, Banking and Finance Department, Kutahya, Turkey. He has been part of the academic staff of Kutahya Dumlupinar University since 2009. He studied Capital Markets at Marmara University and received his bachelor’s degree in 2005. He completed his master’s degree in Financial Economics at Leicester University and received a Ph.D. degree in Business Administration from Kutahya Dumlupinar University. In 2018 at Trier University, he conducted researches on the relationship between R&D, financial structure and institutional ownership of Turkish manufacturing firms. This research was funded by The Scientific and Technological Research Council of Turkey (TUBITAK). His research interest lies in financial literacy, emerging markets, corporate finance, behavioral finance, asset pricing, banking and finance, Islamic banking and finance.