Dynamics of monthly seasonalities in Finnish stock market
Koskinen, Tommi (2004)
Kuvaus
Kokotekstiversiota ei ole saatavissa.
Tiivistelmä
The main purpose of this study is to examine the monthly seasonalities and their dynamics in Finland. Monthly seasonalities are abnormal returns for example in January compared to other months of the year. When investigating the dynamics of the anomaly, the focus is around the fact that abnormal returns differ between years. The data used in this study is the Helsinki Stock Exchange main index data from 1977 to 2002. There are two sets of data, because the main index changed from Unitas-index to HEX-index in 1991. Therefore the data used before 1991 is the Unitas-index and after that the HEX-index is used.
First, the power ratio method, developed by Gu (2002), is used in order to get monthly returns into a measurable form. The power ratio method compares the difference between situations of one month’s return twelve times to years return. This comparison gives an insight of how powerful the possible monthly abnormal is. The goal is to find the months with highest and the lowest abnormal returns. After that a regression analysis is carried out to reveal the relationship between January effect and five explanatory variables, which are GDP growth, inflation, annual return, standard deviation and variance. The focus in regression analysis is not just around January. September and October relationships of the five explanatory variables are also investigated.
The main findings were that the best abnormal return were achieved in January and October, while the worst month was September. These months power ratios had a slightly rising trend in them. The coefficient of variable GDP growth was negative and the only one, which was statistically significant. It was also found that monthly power ratios differ more from each other in years with lower annual return.
First, the power ratio method, developed by Gu (2002), is used in order to get monthly returns into a measurable form. The power ratio method compares the difference between situations of one month’s return twelve times to years return. This comparison gives an insight of how powerful the possible monthly abnormal is. The goal is to find the months with highest and the lowest abnormal returns. After that a regression analysis is carried out to reveal the relationship between January effect and five explanatory variables, which are GDP growth, inflation, annual return, standard deviation and variance. The focus in regression analysis is not just around January. September and October relationships of the five explanatory variables are also investigated.
The main findings were that the best abnormal return were achieved in January and October, while the worst month was September. These months power ratios had a slightly rising trend in them. The coefficient of variable GDP growth was negative and the only one, which was statistically significant. It was also found that monthly power ratios differ more from each other in years with lower annual return.