Dispersion in analysts' forecasts and momentum strategies in the Norwegian stock market

Abstract

Research has shown that stock prices tend to drift in the same direction as revisions in consensus forecasts provided by financial analysts. In this paper we create momentum portfolios by an EPS-earnings revision ratio, and examine raw and abnormal returns for different holding periods for Oslo Stock Exchange (OSE) listed companies in the period 2005-2011. By using the two portfolios with the most and least favourable EPS-revision ratios, a long-short momentum portfolio is created, where we buy the stocks with the most favourable revisions and sell the stocks with the least favourable revisions. We find the ultimate holding period for the portfolios to be three months following the analysts’ forecasts. Our long-short momentum portfolio gives a significant risk free abnormal return of 1% per month. We thereafter introduce the dimension of dispersion in analysts’ forecasts into the analysis, by dividing each portfolio into two sub portfolios by their level of dispersion. The results show that by going long in the sub portfolio with the lowest dispersion and short the sub portfolio with the highest dispersion, we obtain a significant risk free monthly return of 1,33% over the sample period. These findings cannot be explained by classic asset pricing models and contradict the market efficiency hypothesis in its semi-strong form.