Study on Correlation among Oil Prices, Brass Prices and Taiwan Industry Group Stock Price Index

• Main Authors: Hung yue-ren, 黃鉞荏
• Other Authors: Liao Shih-jen
• Format: Others
• Language: zh-TW
• Published: 2007
• Online Access: http://ndltd.ncl.edu.tw/handle/12869892861585598372

碩士 === 樹德科技大學 === 金融保險研究所 === 95 === Energy has been one of the major resources that human beings depend on for survival. It is also one of the main forces that drive economic development of a country. Among all energy supplies, crude oil plays a very significant role. The price increase due to demand is extremely different from the increase in the past three times of global oil crisis due to supply. Earlier issues on oil prices are on correlation with macroeconomicvariables. The author extends the issue on oil prices and common metal (brass) prices analysis across all industries. With the different characteristics of industries in Taiwan, the author further understands the results of different factors. The author adopts Garch model that better captures the data fluctuation of time array to explore the correlation among oil prices, brass prices and Taiwan industry group stock prices in order to understand which group stock prices will be affected by the fluctuation of oil and brass prices. From the evidence results, it is found that the return in one period behind of oil prices is in negative correlation with stock prices of electric machinery, transportation and department stores, meaning that the return in the precedent period of oil prices affects the stock price of electric machinery, transportation and department stores and the fluctuation is very ob-vious as well, causing fluctuation clustering. The return of current and one period behind of brass price is in apparent positive correlation with stock prices of foods, plastics, electric appliances, paper making and steel, etc showing that the stock prices are affected by the return of current and one period behind of brass price and the fluctuation is very obvious as well, causing fluctuation clustering.