The present paper is an empirical test of the Sharpe-Lintner-Mossin model of the CAPM based on a modified version of Gibbons' methodology. We use the rates of return of the data listed in the first section of the Tokyo Stock Exchange. The monthly data on the rates of return from January 1952 to December 1986 was divided into seven five year periods. At 5% significance level the null hypothesis is rejected for all 7 sub-periods in the Alpha sorted portfolios. In the Beta sorted portfolios, the null hypothesis is rejected for three sub-periods. It is concluded that the lack of diversification is the main reason for the invalidity of the CAPM in the Japanese stock market.
ASJC Scopus subject areas