The Stationarity of the Conditional Mean of Real Rates of Return on Common Stocks: An Empirical Investigation

Abstract

The capital asset pricing model of Sharpe [39], Lintner [28], and Mossin [33] has been the basis for many theoretical and empirical studies in capital markets. One criticism of the model hasbeen directed at the assumption that investors optimize in a one-period framework. Fama [7] and Merton [30] have shown that the results of this one-period optimization model are consistent with the results for an intertemporal optimization model if the investment opportunity set is constant over time. Specifically, the relevant parameters for the distributions of risky securities (i.e., conditional means and variances) must be constant over time. Merton has argued that this is a restrictive assumption, but there has been very little empirical evidence to suggest that changes in the investment opportunity set are significant.