Predictability of Stock Returns in a Thin Security Market;
Empirical Evidence with Data from Helsinki Stock Exchange
1975 - 1986
Paavo Yli-Olli and Ilkka Virtanen
Abstract
Stock market efficiency is a crucial concept when forecasting of
future stock returns is discussed. Efficient markets are divided into
three levels of market efficiency depending upon the type of
information used by investors. In a strong form of efficient market
security prices rapidly reflect all information. A semi-strong form
of efficiency respectively holds that no publicly available
information can be successfully used in the prediction of stock
returns. Finally, under a weak form of efficiency, information of
historical stock returns is of no value for predicting the future
stock returns. On a thin security market, like in the Helsinki Stock
Exchange, many anomalies and deviations from market efficiency
have been obtained. It is shown in the paper that the average
Finnish stock market returns do not follow the general random
walk model. Both the monthly and quarterly stock returns can be
forecasted using either univariate time series analysis or
multivariate econometric modelling. Also a procedure is
presented to generate composite forecasts from univariate time
series models and econometric models.
(Vierte Tagung: Geld Banken und Versicherungen)