Forecasting the TAIEX Based on Fuzzy Time Series, Particle Swarm Optimization Techniques and Support Vector Machines

• Main Authors: Pei-yuan Kao, 高培元
• Other Authors: Shyi-ming Chen
• Format: Others
• Language: en_US
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/67t828

碩士 === 國立臺灣科技大學 === 資訊工程系 === 100 === In recent years, fuzzy time series has successfully been used to deal with forecasting problems. From the historical time series data, we can see that the slope of variation of each datum is an important to find the trend of fuzzy time series, where we can use the trend to forecast the variation of future. In this thesis, we present a new method to forecast the Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX) based on fuzzy time series, particle swarm optimization and support vector machines. In the proposed method, we first calculate the slope of change of the daily Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX), and use particle swarm optimization techniques to find optimal intervals in the universe of discourse. Then, we fuzzify the slope of the time series data and generate fuzzy logical relationships to construct fuzzy logical relationship groups. Then, we transfer the forecasting problem into the classification problem and use the support vector machine to deal with the classification. The classified result using the support vector machine is a fuzzy set. Then, we defuzzify the fuzzy set to get the forecasting value of the next trading day. The experimental results show that the proposed method outperforms the existing methods.