Neural Network vs Linear Models of Stock Returns: An Application to the UK and German Stock Market Indicies

We compare the out-of-sample performance of monthly returns forecasts for two indices, namely the FAZ and the FT. A linear and a nonlinear artificial neural network (ANN) model are used to generate out-of-sample competing forecasts for monthly returns. We consider two fundamental variables as the explanatory variables in the linear model and the input variables in the ANN model, namely the trading volume and the dividend. The comparison of out-of-sample forecasts is done on the basis of forecast encompassing. The results suggest that the out-of-sample ANN forecasts encompass linear forecasts of both indices. This finding indicates that the inclusion of nonlinear terms in the relation between stock returns and fundamentals is important in out-of-sample forecasti