We apply the method of constrained asset share estimation (CASE) to test the mean-variance efficiency (MVE) of the stock market. This method allows conditional expected returns to vary in unrestricted ways, given investor preferences. We also allow conditional variances to follow an ARCH process. The data estimate reasonably the coefficient of relative risk aversion, though are unable to reject investor risk neutrality. We reject the restrictions implied by MVE, although changing conditional variances improve statistically upon measured market efficiency. We find that unrestricted asset-share and ARCH models help forecast excess returns. Once MVE is imposed, however, this forecasting ability disappears.