The prediction of Stock Returns in Helsinki Stock Exchange by Abnormal trading volume

Abstract

1. The purpose of this study is to test whether trading volume has any informational role in predicting stock returns. More precisely, the power of trading volume in predicting the direction of future price movements is evaluated. Another purpose is to test whether price movements implied by trading volume shocks are products of the short-and medium-term return autocorrelations.

2. Foreign literature and research studies are used as references. The data consists of all eligible stocks listed on the Helsinki stock exchange during January 1991 to October 2000. Data of weekly returns and trading volumes were collected from the database of University of Vaasa.

3. Sample was splitted into 25 nonintersecting trading intervals of ten weeks. In each trading interval, the first nine weeks were used to measure whether trading volume during the last week (the tenth week) is unusually large or small. Based on this, portfolios were formed at the end of the last week and their performances were evaluated over the subsequent ten weeks.

4. The results showed that stocks with unusually large trading volume will perform better than stocks with unusually low trading volume. After stocks experienced high trading volumes/low trading volumes past losers performed better than past winners. Results were not totally consistent with Gervais, Kaniel and Mingelgrin’s (2001) findings. Gervais et al. (2001) found that stocks experiencing unusually high (low) trading volume over a day or a week tend to appreciate (depreciate) over the course of the following month. The stocks with unusually low trading volume performed better in this thesis than in Gervais et al. (2001) research. Gervais et al. (2001) argued that conditional on trading volume there is no significant relationship between past return and future return, and contemporaneous return and future return. However, I found that earlier returns of security affect strongly to volume based trading strategies.