| At present China's insurance funds are allowed to invest in stocks, securities funds, corporate bonds, government bonds , bank deposits and other assets. How to effectively measure the integrated risk of these different asset portfolio composition is a major problem facing China's insurance industry. Most traditional risk measurement model assumed the joint distribution of the rate of return over the assets is subject to normal distribution, but a large number of empirical studies have shown that this assumption is inconsistent with the objective facts. Therefore under the assumption of normal distribution calculate the portfolio VaR compared to the actual situation is usually have a great deviation.In order to measure the risk of the insurance funds investment portfolio effectively and achieve its optimal configuration, based on China's national conditions, we select the stock index, the fund index, treasury and corporate bonds indices to simulate these assets'investment return. For the defect of traditional risk measurement model, we choose Copula Function to describe the related structures between stocks,funds, treasury bonds and corporate bonds. To reflect the fat-tail, peak and the volatility concentration nature of the distribution of financial assets'return, we established the GARCH(1,1)-t model. Combined with historical data, using Monte Carlo numerical simulation technology generated rate scenario distribution with related structures of Copula function, and then get the value at risk of insurance invest portfolio. Finally we introduce CVaR to replace variance as a tartet function in portfolio optimization, afrer that we established Copula-based insurance portfolio optimization model, and for a given rate of return we get the optimal ratio of investment in China's insurance companies. Insurance companies can adjust their investment structure according to the optimal ratio of investment, to reduce investment risks and improve return on investment. |
PDF Full Text Request