The Case Study of Adaptive Network-Based Fuzzy Inference System Modeling for TAIEX Prediction

摘要

The purpose of this research is to investigate the relation between international stock markets and Taiwan's stock market, and use the statistical method to identify international markets of high correlation with Taiwan's stock market as the input parameters of the ANFIS (Adaptive Network-based Fuzzy Inference System) model to improve the forecasting accuracy. This paper collects dataset in the period of years 2000~2010 from 13 countries including Japan, Singapore, Malaysia of high correlation with Taiwan. to prove the forecasting effectiveness of the model, this paper compares the proposed model with models proposed in other studies. the experimental results suggest that, the total average RMSE of the proposed model is lower as compared with other models. the contribution of this research is to integrate the statistical method and ANFIS model to effectively enhance the forecasting accuracy for TAIEX stock price prediction.