The purpose of this thesis is to examine the causal relationship between the stock index and exchange rate by using Granger causality test in mean and quantiles. To do so, we use the KOSPI stock index and industrial stock index and the won/dollar exchange rate. We use the weekly data, ranging from January 6, 2001 to June 30, 2020.
We first examine the causal relationship between the KOSPI stock index and exchange rate using Granger causality test in conditional mean and Granger causality test in quantiles. Next, we examine the causal relationship between the industrial stock index and exchange rate using the Granger causality test in conditional mean and Granger causality test in quantiles. Finally, we examine the causal relationship between the stock index and the won/dollar exchange rate using the monthly data. The results of the empirical analysis in this thesis are summarized as follows:
First, the causal relationship between the KOSPI stock index and the won/dollar exchange rate using Granger causality test in conditional mean result reports that the KOSPI stock index and the won/dollar exchange rate affect each other.
Second, the quantile causality test shows that the KOSPI stock index Granger cause the won/dollar exchange rate in the low quantiles implying that the result supports the ''portfolio approach''. And the result shows that the KOSPI index and the won/dollar exchange rate affect each other in high quantiles, supporting both the ''goods market hypothesis'' and the ''portfolio approach''.
Third, the empirical results show that the causal relationship between the industrial stock index and the won/dollar exchange rate varies depending on industry. The test result of Granger causality in quantiles indicates that the causal relationship change as the exchange rate and the stock index return fluctuate.
Fourth, the result of the causal relationship between the stock index and the won/dollar exchange rate using the monthly data shows the difference from the empirical result using the weekly data. In particular, we find that the differences in the causal relationship by sub-industries in manufacturing and finance when using the weekly data, while it is observed that there is the similar causal relationship between sub-industries when using the monthly data.
The empirical results of the thesis are expected to be a guide of investment portfolio strategy and show that investor needs to change their investment strategy by industry.