Time dependent Directional Pro t Model for Financial Time SeriesForecastingJingtao

Goodness-of-t is the most popular criterion for neural network time series forecasting. In the context of nancial time series forecasting, we are not only concerned at how good the forecasts t their targets, but we are more interested in proots. In order to increase the forecastability in terms of proot earning, we propose a proot based adjusted weight factor for backpropagation network training. Instead of using the traditional least squares error, we add a factor which contains the proot, direction, and time information to the error function. The results show that this new approach does improve the forecastability of neural network models, for the nancial application domain.