A Research of the Interactive Relationship Among the Exchange rate , the Price of Gold and Crude Oil ---Taiwan Study

碩士 === 真理大學 === 財經研究所 === 100 === This study focuses on the correlation of oil prices, gold prices and NT dollar exchange rate of the U.S. dollar during 2007/09/03 ~ 2011/12/28. Using the cointegration test, vector autoregression (VAR) model, the impulse response, and the variance decomposition...

Full description

Bibliographic Details
Main Authors: Guan-Sian Lee, 李冠賢
Other Authors: Hsiao-Fen Chang
Format: Others
Language:zh-TW
Published: 2012
Online Access:http://ndltd.ncl.edu.tw/handle/61933167651581862145
Description
Summary:碩士 === 真理大學 === 財經研究所 === 100 === This study focuses on the correlation of oil prices, gold prices and NT dollar exchange rate of the U.S. dollar during 2007/09/03 ~ 2011/12/28. Using the cointegration test, vector autoregression (VAR) model, the impulse response, and the variance decomposition to clarify the interactive relationship between the variables. The empirical results of this study are summarized as follows: Variables are stationary in the first order differential of the ADF unit root test; Johansen cointegration test shows no cointegration relationship between these variables. Through the VAR model, the price of gold is more independent, and its impact is the lowest, with oil prices in second place, and exchange rates - the most vulnerable. The Granger causality test shows that the oil price and exchange rate are a two-way feedback relationship, and gold price is one-way leading exchange rate and it is not affected by the other variables. In addition, the impulse response analysis shows that, when interference strikes the three variables, its impulse response effects have maximum reaction in the first day and will disappear after 3 to 4 days later. Also through the variance decomposition, we find that when unexpected changes in the variables’ price occur, gold price can be up to 99% self-explanatory as the exchange rate and oil prices are 93% and 88%.