Stock Market Seasonality Analysis and Comparison of Selected Regions

Abstract

With the development of behavioral finance, scholars around the world have begun to study stock markets in an attempt to find specific patterns that exist. The Efficient Market Hypothesis (EMH) has a very important place in modern financial market investment theory. It was proposed in Fama (1970), while it assumes that the relevant information already available in the securities market is fully utilized and is fully and correctly reflected in the price of securities. However, a large body of empirical research has shown that there are anomalies in the stock markets, the occurrence of extraordinary returns on certain assets or portfolios of assets, which cannot be explained by the efficient market hypothesis. One of the important anomalies in the stock markets is the calendar effect, which refers to the phenomenon of extraordinary returns that can be achieved by trading at a particular time in the stock markets. There are many types of calendar effects, including day-of-the-week, month-of-the-year, holiday, and seasonal effects. These anomalies may not be always disadvantageous for investors, and it may even be possible that investors who follow these effects may earn some unusual income. The aim of this thesis is to find out whether there are significant differences in calendar effect within different geographical regions. Five stock indices, each representing one geographical area were selected to achieve the aim. Using the research methodology of a GARCH model, a computational analysis is carried out to determine when the weekly and monthly effects of each of the five selected stock markets are present.

With the development of behavioral finance, scholars around the world have begun to study stock markets in an attempt to find specific patterns that exist. The Efficient Market Hypothesis (EMH) has a very important place in modern financial market investment theory. It was proposed in Fama (1970), while it assumes that the relevant information already available in the securities market is fully utilized and is fully and correctly reflected in the price of securities. However, a large body of empirical research has shown that there are anomalies in the stock markets, the occurrence of extraordinary returns on certain assets or portfolios of assets, which cannot be explained by the efficient market hypothesis. One of the important anomalies in the stock markets is the calendar effect, which refers to the phenomenon of extraordinary returns that can be achieved by trading at a particular time in the stock markets. There are many types of calendar effects, including day-of-the-week, month-of-the-year, holiday, and seasonal effects. These anomalies may not be always disadvantageous for investors, and it may even be possible that investors who follow these effects may earn some unusual income. The aim of this thesis is to find out whether there are significant differences in calendar effect within different geographical regions. Five stock indices, each representing one geographical area were selected to achieve the aim. Using the research methodology of a GARCH model, a computational analysis is carried out to determine when the weekly and monthly effects of each of the five selected stock markets are present.