An Investigation of the Day-of-the-Week Effect on the Istanbul Stock Exchange of Turkey

INTRODUCTION Until about fifteen-twenty years ago, empirical evidence has steadfastly indicated that equity market returns depend on day of week. For U.S. market indicators, documented evidence had suggested a significantly negative and lowest rate of return on Mondays and a diametrically opposite pattern on last trading day of week. A comparable evidence of day-of-the-week anomaly has also been documented for both developed and developing markets around world. Kamath, et al. paper (1998) presents a table of empirical findings regarding this anomalous in Australia, Canada, Finland, France, Germany, Greece, Italy, Netherlands, Spain, Sweden, Switzerland, United Kingdom, Hong Kong, Japan, Korea, Malaysia, New Zealand, Philippines, Singapore, and Thailand. More recently, some researchers have questioned findings reported by utilizing OLS methodology (see for example Connolly (1989), Chang, et al. (1993) and Dubois and Louvet (1996)). However, reliance on a robust econometric technique to observe day-of-the-week has not always succeeded in disputing OLS findings. For example, while Alexakis and Xanthakis (1995), and Kamath, et al. (1998) papers concluded that evidence on anomalous was not dependent on methodology, Chen, et al. (2001) study noted that their findings of Chinese stock market were dependent on estimation method as well as period of study. The purpose of this study is to present empirical evidence on day-of-the-week on Istanbul Stock Exchange of Turkey. The secondary objective is to determine if conclusions regarding presence of such an are dependent on methodology adopted as well as sample period itself. To meet these objectives, we rely on most recent 60-month period from January 2003 through December 2007. The rest of paper is organized as follows. A brief review of literature on subject at hand is contained in next section. The data and methodologies utilized are described in third section. The findings of this investigation are reported in fourth section. A brief summary of paper is included in last section. PREVIOUS RESEARCH Abundant empirical evidence concerning day-of-the-week as well as other seasonal anomalies in equity markets can be found in finance literature. Specifically, some of most cited articles on subject (French (1980), Gibbons and Hess (1981), Keim and Stambaugh (1984), Jaffe and Westerfield (1985) and Lakonishok and Smidt (1988)) have noted that market indicator returns tend to be dependent on day of week. For example, Lakonishok and Smidt (1988) have reported that such on anomaly existed in each of their ten sub-periods dating back to 1897 for Dow Jones. By and large, methodology of choice in earlier studies was standard OLS formulation. Connolly (1989) questioned evidence of such an based on OLS methodology since he believed that distributional attributes of equity market data were inconsistent with underlying assumptions of OLS technique. Connolly also noted that interpretation of findings of studies which relied on very large sample sizes could have been distorted. In particular, he indicated that in such cases, F-tests tend to reject null hypothesis too often. Connolly's study concluded that intensity of day-of-the-week in U.S. markets had reduced considerably after 1975. Indeed studies of Keim and Stambaugh (1984), Rogalski (1984), Smirlock and Starks (1986), and Condoyanni, et al. (1987) detected reduced intensity of such an in post-1975 sub-periods of their respective studies. Chang, et al. (1993) study confirmed that the effect had become insignificant in U.S. as well as in Belgium, Denmark and Germany during 1986-1992 period upon adjusting for OLS error term or sample size. …

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.