How Do Neural Networks Enhance the Predictability of Central European

In this paper, we apply neural networks as nonparametric and nonlinear methods to Cen-tral European (Czech, Polish, Hungarian, and German) stock market returns modeling. In the first part, we present the intuition of neural networks and we also discuss statisti-cal methods for comparing predictive accuracy, as well as economic significance mea-sures. In the empirical tests, we use data on the daily and weekly returns of the PX-50, BUX, WIG, and DAX stock exchange indices for the 2000–2006 period. We find neu-ral networks to have a significantly lower prediction error than the classical models for the daily DAX series and the weekly PX-50 and BUX series. We also achieve economic significance of the predictions for both the daily and weekly PX-50, BUX, and DAX, with a 60 % prediction accuracy. 1