A hybrid arima and neural network for yields prediction

Abstract

The accuracy of time series forecasting is fundamental to many decisions processes and hence the research for improving the effectiveness of forecasting models has never been stopped (Zhang, 2003). Recent research activities in artificial neural network (ANN) have shown powerful pattern classification and pattern recognition capabilities. One major application area of ANN is forecasting (Sharda, 1994). ANN provides an attractive alternative tool for both forecasting researchers and has shown their nonlinear modeling capability in data time series forecasting.