Modelling mean reversion of asset prices towards their fundamental value
将基本面价值建模为随机过程，资产价格向其均值回归，用盈利和股息作为代理变量进行实证，模型解释力良好，与Campbell-Shiller和Fama-French模型比较表现相当或更优。
In this paper we extend the study of mean reversion behavior by modelling the fundamental value as a stochastic process. The market value of the asset is then modelled as a mean reverting Ornstein Uhlenbeck process towards the fundamental value. Solving backwards, we determine the functional form of the regression equation of changes in asset prices and returns to changes to the fundamental value. Using earnings and dividends as proxies for the fundamental value we test our model empirically. In general, other than the shortest horizon of 1-year, our model shows good explanatory power. Since our model is compatible with Campbell and Shiller (1988) framework in the earnings case and Fama and French (1988) model in the dividend case, the performance of our model has been compared with those two models. In comparison, the performance of our model is comparable to that of Campbell and Shiller and compares favorably with Fama and French.