Volatility Forecasting of USD/NTD Exchange Rate and Its Relationship with Forward Exchange Rate: Effects of Forecasting Performance and Trading Volume

• Main Authors: Tze-Chen Yang, 楊慈珍
• Other Authors: Hsiang-Hsi Liu
• Format: Others
• Language: zh-TW
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/65789549836510718269

碩士 === 國立臺灣海洋大學 === 應用經濟研究所 === 94 === Abstract The purpose of this study is to establish the relationships between the spot and forward exchange rate and to forecast the volatility of both USD/NTD exchange rates. This study applies the following six single variate models, such as stochastic volatility model, GARCH model, GARCH-M model, EGARCH model, TGARCH model and GJR-GARCH model, to forecast the volatility of the return rate for both spot and forward exchange rate. Comparing the forecasting performance of the above six models, the VEC-TGARCH model is chosen to specify the bi-variates relationship between the spot and forward exchange markets. The sample period is from January 2, 2001 to November 30, 2005. Major conclusions of this study are shown as follows. First, the result of the unit root test shows that both of the USD/NTD spot exchange rate and the USD/NTD forward exchange rate are non-stationary series and are integrated order one. Second, by using Johansen co-integration test, there is a single co-integration relationship between the spot and forward exchange markets. Third, there exists the volatility clustering phenomenon and an asymmetric effect in the spot and forward exchange markets. Fourth, after taking the return of the forward exchange rate and the trading volume into account, the volatility clustering effect will be reduced and the forecasting of the volatility will perform better. Fifth, there exists reciprocal cause and effect relationship between spot and forward exchange markets and the reaction of the forward exchange market to any new intervention is larger than that of the spot exchange market. Sixth, by comparing the forecasting performance of the volatility from the above six models metioned and the VEC-TGARCH model, the stochastic volatility model ranked the best, the VEC-TGARCH model ranked the second, and the TGARCH model ranked the third.