The Study of Return of Vector Stock Forecasting model－The Application of 3SLS and SSM

• Main Authors: Nanchi Tseng, 曾南溪
• Other Authors: GOO, YEONG-JIA
• Format: Others
• Language: zh-TW
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/65538626898728009897

碩士 === 國立臺北大學 === 企業管理學系 === 94 === This article uses the macroeconomic variables to establish forecasting models. The macroeconomic variables include interest rate, exchange rate, money supply, unemployed rate, price index, Taiwan stock index, foreign stock index. The research period is from January, 1995 to December, 2005. The steps of this research: 1.Transforming the raw data. 2. Using the stepwise regression to chose the variables which affect return of industrial stock. 3. Using the effective variables to establish the 3LS model. 4. Using different return of industrial stock variables to establish the SSM. 5. Comparing the forecasting performance of 3SLS and SSM. The empirical results can be summarized as follow: 1. Significant variable to forecast industrial stock index include CPI, USD-NTD exchange rate, S&P500 index. 2. To forecast Financial industry index and Steel industrial index: Because the variables which are used to forecast industrial stock index by 3SLS not only consists of return of industrial stock index but also macroeconomic variables, the forecasting performance of 3SLS is better than SSM. 3. To forecast Electronics industry index, Petrochemistry industrial index and Transportation industry index: Because SSM considers the interactive relationship of different industrial stock index, the forecasting performance of SSM is better than 3SLS. 4. Because the variables which are used to forecast industrial index by 3SLS not only consists of return of industrial index but also macroeconomic variables, the average forecasting performance of 3SLS is better than SSM.