A reserach on the relationship among accommodations with collateral rate of the central bank, 3 month U.S. dollar deposit rate and 90 day U.S. treasury bill rate:plus GARCH model test

• Main Author: 葉家全
• Other Authors: 廖世仁
• Format: Others
• Language: zh-TW
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/40490786711037130981

碩士 === 樹德科技大學 === 金融保險研究所 === 91 === Abstract This paper discussed the relationship among Accommodations With Collateral Rate of the central bank, 3-month U.S. Dollar Deposit Rate and 90-day U.S. Treasury Bill Rate ; secondly, it further analyzed fluctuating ratio of these three variables, and it also compared the difference between the demonstration results of Granger Causality and Vector Error Correction model. Additionally, it analyzed their relations using Vector Auto Regression against the fluctuating ratio of these three variables. The study period of this paper was from Aug. 1985 to May 2002, and the data were mainly from Aremos Economic & Statistic Database of the Computer Center of the Education Ministry and U.S. Central Bank Database. The demonstration results showed that these three variables displayed stationary states after first order difference. For the aspect of cointegration test, it found long-term equilibrium relationship among them. While Granger Causality and Vector Error Correction model found that different interaction was existed among these variables. For the aspect of fluctuation ratio, these variables themselves had stationary state. For the aspect of Vector Auto Regression, there was different interaction among them. Examined by GARCH model, this paper found that there was GARCH (1,1) and volatility clustering among Accommodations With Collateral Rate of the central bank, 3-month U.S. Dollar Deposit Rate and 90-day U.S. Treasury Bill Rate . Finally, Variance Decomposition verified how much the difference of fluctuating ratio of variables was influenced by the variables. The demonstration results found that the variation of each variable were mainly influenced by their own interaction.