An Empirical Study of Value at Risk Measurement Model --The Case of Agribusiness Portfolio

• Main Authors: Yu-Fang Weng, 翁玉芳
• Other Authors: Rern-Jay Hung
• Language: zh-TW
• Published: 2002
• Online Access: http://ndltd.ncl.edu.tw/handle/91481016688802234728

碩士 === 國立屏東科技大學 === 農企業管理系 === 90 === In history of Taiwan agriculture development, agriculture industry declined even though it had a prosperous period in past days. With the change of the economic style, agribusiness has replaced the traditional agriculture and played a predominant role in agriculture management recently. At the present day, most of the listed and over-the-counter agribusiness companies in Taiwan are food industries. Due to the low stock the prices of the listed food companies in the past few years, government’s encouragement to invest in agriculture biotechnology, and the soaring price of the securities in biotechnology industry in USA, many of the food companies joined the agriculture biotechnology research, and investors hold positive viewpoints in biotechnology and the growth of the securities in agribusiness in the future. Therefore, we can expect that the number of the investors will increase to invest relevant agribusiness securities in the future. Thus the risk which investors may encounter in the market becomes a crucial subject. In order to comprehend the risk condition that agribusiness investors may face, the study constructed an Agribusiness Portfolio as a study subject and used the most accepted technique — Value at Risk technique — to measure the value of risk. The risk value of the Agribusiness Portfolio was estimated by three of the most used models — Variance/Covariance, Historical Simulation Method, and Monte Carlo Simulation. By using the estimated abilities of three models, we attempt to find out the optimal measured one to measure the risk value of Agribusiness Portfolio so that we can provide the reference of the risk management to the Agribusiness Portfolio investors. We can get the best-estimated result in Standard Variance Method by using GARCH Model, and get best-estimated one in Percentile Method by using Historical Simulation Method Model. In general, we can get the best-estimated result by using Historical Simulation Method Model, since that it is suitable to measure the Value at Risk of Agribusiness Portfolio by the above model. According to the conventional estimation, one holds one million dollars in Agribusiness Portfolio will bear 62,214.24 to 62,629.24 dollars extreme loss daily under 99% confidence level from Jan. 2, 2001 to Dec. 31, 2001. After comparing with the other studies, we can find that the risk of the Agribusiness Portfolio is higher than that of the portfolio of the financial related stocks, but less than the risk of the portfolio of the electronic related stocks. The risk of Agribusiness Portfolio is between the former two portfolios.