论文部分内容阅读
为了准确地量化资产之间的时变相依结构和预测组合风险,本文考虑到投资者对资产风险偏好的差异,假设资产收益率序列的新息服从标准t分布,提出时变Copula-GARCH-M-t模型,推导了模型参数的两步MCMC估计方法,还得到了组合风险(VaR和CVaR)的一步预测方法。最后选取上证综合指数和标准普尔500指数,验证了所提模型及方法的可行性和优越性,同时该模型较为准确地量化了两指数在次贷危机后的时变相依结构特征。
In order to accurately quantify the time-dependent dependency between assets and predict the portfolio risk, this paper takes into account the investor’s preference for asset risk. Assuming that the interest rate of the asset return series follows the standard t distribution, the paper proposes time-varying Copula-GARCH-Mt Model, two-step MCMC estimation method of model parameters is deduced, and a one-step prediction method of portfolio risk (VaR and CVaR) is also obtained. Finally, we choose the Shanghai Composite Index and the S & P 500 index to verify the feasibility and superiority of the proposed models and methods. Meanwhile, the model more accurately quantifies the time-dependent dependence of the two indices after the sub-prime crisis.