The primary objective of this paper is to derive a multi-factor equilibrium model using a mean-variance approach. The results of this derivation provide greater insight into the nature of the resulting factors than does APT. There are several important implications for empirical tests of any a priori defined multi-factor model.

This paper examines the effect of risk differences on the oft-documented negative rela? tionship between stock returns and inflation. We find risk-related patterns of coefficients on our estimates of the level and change in expected inflation and on unexpected inflation. These patterns are consistent with the hypothesis developed in Fama [2] and in Geske and Roll [7] that future real output growth simultaneously helps to determine current stock returns and various measures of inflation.