Forecating S&P 500 index by Weighted Average Defuzzification Based on NEWFM

Fuzzy neural networks have been successfully applied to generate predictive rules for stocks forecasting. This paper presents a methodology for forecasting S&P 500 index based on the neural network with weighted fuzzy membership functions (NEWFM) and time series of S&P 500 index based on the defuzzyfication of weighted average method (The fuzzy model suggested by Takagi and Sugeno in 1985). NEWFM is a new model of neural networks to improve forecasting accuracy by using self adaptive weighted fuzzy membership functions. The degree of classification intensity is obtained by bounded sum of weighted fuzzy membership functions extracted by NEWFM, and then weighted average defuzzification is used for forecasting S&P 500 index. In this paper, the Haar wavelet function is used as a mother wavelet. A set of two extracted coefficient features and a set of two extracted approximation features are presented to forecast S&P 500 index using the Haar WT. About 90% of the data, from Oct 1, 1990 to Oct 1, 2002 is used for training and 10% for testing. The result of classification rate is 54.1%. The implementation of the NEWFM demonstrates an excellent capability in the field of stocks forecasting.