Abstract
Recent empirical studies have found that stock returns contain substantial negative serial correlation at long horizons. We examine this finding with a series of Monte Carlo simulations in order to demonstrate that it is consistent with an equilibrium model of asset pricing. When investors display only a moderate degree of risk aversion, commonly used measures of mean reversion in stock prices calculated from actual returns data nearly always lie within a 60 percent confidence interval of the median of the Monte Carlo distributions. From this evidence, we conclude that the degree of serial correlation in the data could plausibly have been generated by our model.