Using Clustering Methods to Explore the Consistency between Exchange Rate and Stock Index

• Main Authors: WU, MEI-HUI, 吳美慧
• Other Authors: YU, CHANG-YUNG
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/55726872737160940241

碩士 === 靜宜大學 === 財務與計算數學系 === 104 === In this study, we discuss the relationship between exchange rates and stock indices. The exchange rates during the period of time from 1990 to 2016 are collected and cluster analysis is applied. With defined similarity, clustering technique can group objects based on the principle of maximizing the intra-class similarity and minimizing the inter-class similarity. In this study, we use correlation as distance and similarity between these countries’ exchange rate is observed. Since the relationship between these countries’ exchange rate may vary with time, we separate 1990 to 2016 into six time intervals and the cluster analysis is applied for each time interval data. The group results show that correlation between countries’ exchange rate did change at different time intervals. Similar analysis is applied to stock indices of these countries. Also, the grouping structure change with time. The results are also used to explore the relevance between exchange rate and stock indices. Only the European countries are highly consistent for both exchange rate and stock indices over all time intervals. At last, the regression analysis is applied to observe the linear relationship between exchange rate and stock indices for each individual country and the most of the results show no significant linear relationship results between exchange rate and stock price index of each country.