| Summary: | 碩士 === 開南大學 === 企業與創業管理學系 === 96 === This paper using Cointegration, Vector Autoregression Model, Granger Causality test, Impulse Response Function and Forecast Error Variance Decomposition , to aimed at the dynamic relevant of Taiwan’s house price, stock price, interest rates and exchange rate, which the data was adopted from Sep/1991 to Dec/2007. The conclusions as following：
1.The conclusion of Johansen Cointegration : The house price in Taipei City, Taipei County and Kaohsiung City were kept long term steady and balance relation with the stock prices, interest rates and exchange rate.
2.The conclusion of Granger Causality testing: There was two-way feedback relation between house price and stock price in Taipei city and Taipei County. Stock price and interest rate were leaded house price in Taichung city. Exchange rate was leaded house price in Taipei City and Kaohsiung city. All of them were one-way causality. Another test conclusion of regional house price, the variation order of the house price was: (1) Taichung city; (2) Taipei county and Taiwan area (It’s no causality to each other) ; (3) Taipei city house; (4) Kaohsjung city. The house price of Taichung can be an indicator for Taiwan’s house price. Those conclusions can be used for predict regional house price variation which people can make the best investment decision or policy created by government.
3.The conclusion of Impulse Response Function and Forecast Error Variance Decomposition：In comparison of each regional house price self explanation capacity, the highest one was Kaohsiung city, the others in ordering were Taipei County, Taiwan area, Taipei city and Taichung city. The highest impact factor of house price in Taipei City and Taiwan area was stock price impact. The highest impact factor of house price in Taichung was interest rate. The highest impact factor of house price in Kaohsjung city was exchange rate.
|
|---|
