Using the Public Information and the Logistic Regression model to Predict the Changes of Taiwan Stock Market Index

• Main Authors: Yi-Ching Li, 李怡青
• Other Authors: Chun-Nan Chen
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/63232058487691200244

碩士 === 國立成功大學 === 財務金融研究所 === 93 === 　The study is to use the daily public information provided by mass media and to apply logistic regression to forecast the changes of Taiwan stock market index. The sample period covers form January 1, 1999 to December 31, 2004. Empirical results indicate that: (1)The significant variables in the open index predicting model are trading volume of foreign investors divided by trading volume of the market, the growth rate of SIMEX index , the growth rate of spread of SIMEX index futures, the spread of SIMEX index futres divided by SIMEX index futures close price, the growth rate of margin trading, the growth rate of Nasdaq index ,the growth rate of Dow Jones index and the growth rate of TSMC's ADR. The percentage of correctness is 80%. (2) The significant variables in the close index predicting model are the change of [(bid volume-ask volume)/ (bid volume+ ask volume)], (bid orders-ask orders)/(bid orders+ ask orders), (average bid volume-average ask volume)/ (average bid volume+ average ask volume), the change of(volume of margin trading /trading volume of the market), the growth rate of Nasdaq index, (today’s open index-yesterday’s close index)/yesterday’s close index.The percentage of correctness is 66%.