CALENDAR ANOMALIES IN THE HELSINKI STOCK EXCHANGE
Kivikari, Jani (2008)
2008
Kuvaus
Opinnäytetyö kokotekstinä PDF-muodossa.
Tiivistelmä
Market efficiency and anomalies have been in interest of academic research over 40 years. Study of calendar effects is also fairly old branch and it can be traced back to 1970’s. Due to long history of studies in the area, evidence of these phenomena is no less than exhaustive. Most of the previous studies are conducted using UK or US data. This study investigates the existence of the most known calendar effects in a thin securities market; the Helsinki Stock Exchange.
In this study the Efficient Market Hypothesis is tested as its weak form. Thus the main question includes the predictability of future returns by analysing past returns. Patterns in stock returns are analysed by regression model. The impact of company size and market direction are analysed by data selection. Data of this study includes returns of five Finnish stock indices and an individual share of Nokia Corporation. Selection of indices is based on how well they indicate returns of different sized companies. The whole data period from 1992 to 2008 is divided into four sub periods in order to find differences in the existence of calendar effects during different market situations and analyse the persistence of these phenomena.
This study provides evidence of weak month-of-the-year effect and the-week-of-the-month effect in the Helsinki Stock Exchange. Further on, no evidence of the existence of the-day-of-the-week effect was detected. Holiday effect, according to results, seemed to be the strongest calendar anomaly in the Helsinki Stock Exchange. Also notion of firm size affecting to calendar effect was made. Market direction also seemed to have an impact in calendar effect. Exception was the holiday effect and especially prior Independence Day and Christmas which were strong throughout the sub periods and thus throughout different market situations.
Some of the most known calendar effects exist in the Finnish stock markets. Even if these effects are not strong enough to enable arbitrage, they could be good guidelines for investors to find an optimal timing for their investment decisions.
In this study the Efficient Market Hypothesis is tested as its weak form. Thus the main question includes the predictability of future returns by analysing past returns. Patterns in stock returns are analysed by regression model. The impact of company size and market direction are analysed by data selection. Data of this study includes returns of five Finnish stock indices and an individual share of Nokia Corporation. Selection of indices is based on how well they indicate returns of different sized companies. The whole data period from 1992 to 2008 is divided into four sub periods in order to find differences in the existence of calendar effects during different market situations and analyse the persistence of these phenomena.
This study provides evidence of weak month-of-the-year effect and the-week-of-the-month effect in the Helsinki Stock Exchange. Further on, no evidence of the existence of the-day-of-the-week effect was detected. Holiday effect, according to results, seemed to be the strongest calendar anomaly in the Helsinki Stock Exchange. Also notion of firm size affecting to calendar effect was made. Market direction also seemed to have an impact in calendar effect. Exception was the holiday effect and especially prior Independence Day and Christmas which were strong throughout the sub periods and thus throughout different market situations.
Some of the most known calendar effects exist in the Finnish stock markets. Even if these effects are not strong enough to enable arbitrage, they could be good guidelines for investors to find an optimal timing for their investment decisions.