Cross Sectional Analysis of the Swedish Stock Market

This paper analyses the ability of beta and other factors, like firm size and book-to-market, to explain cross-sectional variation in average stock returns on the Swedish stock market for the period 1980-1990. We correct for errors in variables problem of the estimated market beta. Since this method takes into account the measurement error we do not have to form portfolios and thereby losing information. We use both separate cross-sectional regressions and a pooled regression model to estimate the risk premiums of the different factors. An Extreme Bounds Analysis is utilised for testing the sensitivity of the estimated coefficients to changes in the set of the included explanatory variables. Since the tests are carried out on realised returns, which presumably are quite noisy approximations of expected returns, we study if beta can systematically explain cross-sectional differences among realised stock returns conditional on the sign of the realised market excess return. Our results show that the coefficient for beta is never significantly different from zero, but the estimates differ across the methods mentioned above. However, we find that beta is priced differently in periods with positive versus periods with negative realised market return. In the Extreme Bounds Analysis, the coefficient for the size variable is always significantly negative.