基于贝叶斯方法与时变Copula模型的基金风险的度量
Measurement of Fund Risk Based on Bayesian Method and Time-varying Copula Model

基于贝叶斯理论的MCMC方法对单个基金收益率进行GARCH建模，以及对投资组合权重进行后验模拟。进一步结合时变Copula理论计算基金投资组合的VaR,与基于极大似然法的结果进行比较。实证结果表明基于贝叶斯理论的时变Copula的VaR方法，能够更有效的度量开放式基金投资组合的风险。
The MCMC method based on Bayesian theory is used to carry out the GARCH modeling on the return of single fund and the portfolio weights posterior simulation. And then the portfolio's VaRs are calculated based both on the time-varying Copula, and maximum likelihood method for comparison. The empirical results show that the measurement based on Bayesian Copula to measure the risk of open-end fund investment portfolio is more effectively.