Cross-sectional estimation of stock returns in small markets: The case of the Athens Stock Exchange

Abstract

This study is an investigation into the cross-sectional determinants of stock returns in a small market - the Athens Stock Exchange - where the Fama and French portfolio grouping procedure that is normally used to counter the error in variables problem in estimating beta is problematic due to the small number of stocks. A maximum likelihood technique is applied, similar to that developed by Litzenberger and Ramaswamy (Journal of Financial Economics, 7, 163–95, 1979), which is arguably a better procedure than the portfolio grouping method even for investigating large (developed) markets. A further empirical problem that was addressed was the possibility that the results were being driven by the ‘January effect’. The findings for the Athens market suggest that there is only one substantive variable in explaining the cross-sectional variation of market and that is market equity ME (which captures a size effect).