The Prediction of Taiwan Stock Index-Taking Electronic Index as an example

碩士 === 輔仁大學 === 應用統計學研究所 === 93 === The thesis takes Electronic Index of the integrated market in Taiwan securities as the study subject. The span of the study ranges from August 1, 1995 to December 31, 2004. The research processes the analysis of three independent variables (NASDAQ Index, PSE Ind...

Full description

Bibliographic Details
Main Authors: Shun-Hung Yang, 楊舜閎
Other Authors: Deng-Yuan Huang
Format: Others
Language:zh-TW
Published: 2005
Online Access:http://ndltd.ncl.edu.tw/handle/39165674248587436546
Description
Summary:碩士 === 輔仁大學 === 應用統計學研究所 === 93 === The thesis takes Electronic Index of the integrated market in Taiwan securities as the study subject. The span of the study ranges from August 1, 1995 to December 31, 2004. The research processes the analysis of three independent variables (NASDAQ Index, PSE Index, and the Exchange Rate from NTD to USD) and the dependent variable (Electronic Index of the integrated market). The study methods of the thesis include Descriptive Statistics Analysis , Correlation Analysis, CART Analysis, and Artificial-Neural-Network Analysis. To understand the index changes before and after the political-party alternation, the writer divides the study range into two sections: the early stage of the study and the late stage of the study. The temporal watershed is May 20, 2000. The study presents the following findings: 1. In terms of the average single-day rise in recent 10 years, Taiwan’s Electronic Index performs the best. SOX Index fluctuates the most. 2.The average single-day reward rate of Taiwan Electronic Stocks falls the most seriously after the political-party alternation. 3.The investment risk of Electronic Index falls on the late stage of the study. 4. NASDAQ Index, the rise and fall of PSE Index, and those of Electronic Index prove positively related. The exchange rate from NTD to USD and the rise in Electronic Index appear positively related. 5.The correlation between PSE Index and Electronic Index is the closest. 6.Three independent variables are all important factors influencing the rise and fall of Electronic Index. 7.The influences of NASDAQ Index and the exchange rate from NTD to USD are more obvious on the early stage of the study while the influence of PSE Index is more prominent on the late stage of the study. 8.Using Artificial-Neural-Network to predict the Electronic Index of the integrated market acquires rather good predictive results.