碩士 / 淡江大學 / 財務金融學系碩士在職專班 / 98 / This empirical study conducts nonlinear research method in smooth transition autoregressive model (STAR) to explore the correlation between the influential international stock index and TAIEX , adopting the countries stock indices from US , China and Hong Kong in this study, and taking the average volatility of these three countries stock indices as explanatory variables.
The evidence finds that the transitional variable is one period lagged and the exponential smooth transition model is specified for our examination. The result shows there exists one threshold value of the average volatility of these three countries stock indices for the effect of major country’s stock return on Taiwan index return. When the index average volatility is far greater or less than the threshold value, the effect of the explanatory variables influenced on the TAIEX is closely two times the value as index average volatility is equal to the threshold value.
Since there is a time difference between Taiwan and US, a one-period (one-day) lagged adjustment of US S&P500 stock index return is necessary. In comparison with previous evidence, the logistic smooth transition model is specified in this lagged adjustment examination and transitional variable is found to be eight period lagged. The result again finds one threshold value in our adjustment model. The final finding is that US, China and Hong Kong’s stock index returns have positive effect on TAIEX return. Furthermore, the influence is much greater as the three countries stock index average volatility is less than threshold value.
We conclude that the examination is more precise with one period lagged adjustment, and the three major international stock index returns have positive effect on TAIEX return.
| Identifer | oai:union.ndltd.org:TW/098TKU05304002 |
| Date | January 2010 |
| Creators | TE-KANG LAO, 勞德康 |
| Contributors | Chien-Chung Nieh, 聶建中 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 56 |
Page generated in 0.1368 seconds