Summary: | 碩士 === 大葉大學 === 事業經營研究所 === 91 === Abstract
The Taiwan Futures Exchange launched its mini Taiwan stock index futures (MTX) on April 9, 2001 and made Taiwan’s financial market much more diversified. Although the MTX was launched later than other index futures, it is also the most popular one.
The main purpose of this thesis is to combine the gray relational analysis and the correlation analysis with neural network to build a prediction model for MTX. The research period is from Apr. 9, 2001 to Jan. 15, 2003. Four main input variables include the original data, technical indicators, Taiwan Stock Exchange Index and foreign stock market index.
The empirical results show that, applying the input variables filtered by the gray relational analysis and the correlation analysis, the rate of the prediction accuracy reaches 90 percent plus. The prediction results are robust to the different input variables. Thus, we conclude that this model is efficient.
|