欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
本文假设单变量时序的新息服从标准的学生t分布，提出多元时变Copula—GARCH—t模型，利用蒙特卡洛马尔科夫链（MCMC）算法对模型参数进行贝叶斯统计推断，给出了多个资产组合风险VaR和CV水的度量方法，并基于风险最小化原则确立了最佳的资产配置模型。实证分析表明，MCMC方法优于经典的IFM方法，能够充分捕捉到中关股市的时变相依结构及相关系数和尾部指数的动态特征。
英文摘要：
The article proposes a time - varying Copula- GARCH - t model, which assumes that the innovation comes from the standard students＇ t- distribution. By using the MCMC algorithm to a Bayesian statistical inference, we obtain onestep CVaR and VaR prediction method, and further establish an optimal weight of asset allocation based on risk minimization principle. For the analysis of Shanghai composite index, Hang Sang Index, TWII and America＇s S＆P 500 , the results show that, MCMC method is superior to the classical IFM method, which can fully capture the time - varying dependency structure and the dynamic characteristics of the correlation coefficient and tail index of the stock markets during the financial crisis.
同期刊论文项目
同项目期刊论文
期刊信息
位置：