An Application of Copula-GARCH on Asset Allocation: A Case for Gold, Oil, Cotton, Stock, and Bond

• Main Authors: Chun-Hsiang Chang, 張峻祥
• Other Authors: Wang Chou Wen
• Format: Others
• Language: zh-TW
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/56etv2

碩士 === 國立中山大學 === 財務管理學系研究所 === 101 === This paper applied Copula-GARCH methodology for asset allocation of a portfolio with commodities, including, gold, oil, cotton, stock, and bond. We used GARCH(1,1)-student- t to fit the marginal distribution. Instead of correlation, we applied Copula to capture the dependence structure between assets, and solved the optimal weight by minimizing CVaR, standard deviation, or maxing Mean-Variance, or CRRA utility function. In this study, we found that it is optimal to invest in future 10 days based on the historical data of the past 126 days. We constructed portfolios for investors with different degree of risk aversion. The empirical results showed that the less risk aversive, the higher both portfolio performance and volatility is. Finally, we observed the change of weights between assets. The stock primarily constituted the portfolio before 2008. During the period of financial tsunami, the proportion of bond and gold grew up significantly. After 2009, commodities played an important role in portfolio.