A Study on the Predicting Trend of Taiwan Stock Index Futures Using Decision Trees and Artificial Neural Networks

• Main Author: 林俊騰
• Other Authors: 黃華山
• Format: Others
• Language: zh-TW
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/18167903610641718800

碩士 === 國立彰化師範大學 === 資訊管理學系所 === 99 === Some people regard the Taiwan index future as a hedge tool or an investment target with high leverage in the futures market. Every technical indicator has its own fanciers. However, the purposes of investment behaviors are conferring which technical indicators can forecast future trends instead of well using. Therefore, the study use 23 kinds of technical indicators and establish 3 molds by using decision tree, neural networks and combination both. We expect to find out the important variables which could forecast the next day’s price fluctuation and build up a model with high predicting ability. The study adopts the data of Taiwan Index Future for 12 with total amount of 3164 which will be separated into training set (2335) and testing set (829). We also divided into market tick, up and down to build model. The study result shows the most powerful model is using combined type which the accuracy rate is 85.89%. On the investment point, model 3 can actually predicate the up and down of next day in this three years.