Predicting the UK Equity Premium with Dividend Ratios: An Out-Of-Sample Recursive Residuals Graphical Approach

The purpose of this paper is to evaluate the ability of dividend ratios to predict the UK equity premium. Specifically, we apply the Goyal and Welch (2003) methodology to equity premia derived from the UK FTSE All-Share index. This approach provides a powerful graphical diagnostic for predictive ability. Preliminary in-sample univariate regressions reveal that the UK equity premium contains an element of predictability. Moreover, out-of-sample the considered models outperform the historical moving average. In contrast to similar work on the US, the graphical diagnostic then indicates that dividend ratios are useful predictors of excess returns. Finally, Campbell and Shiller (1988) identities are employed to account for the time-varying properties of the dividend yield and dividend growth processes. It is shown that by instrumenting the models with the identities, forecasting ability can be improved.