论文部分内容阅读
对由上证综合指数、深证成分指数、上证基金指数、上证国债指数计算的日自然对数收益率组成的数据矩阵,分别建立了残差服从正态分布、t分布的向量ARCH、向量GARCH、纯对角GARCH、BEKK、常条件相关GARCH、主成分GARCH和EWMA模型,基于这些模型,计算了风险价值(VaR),进而通过比较计算结果,得出BEKK—t模型测算中国金融市场投资组合的风险价值(VaR)效果最好等的结论.
For the data matrix consisting of the daily natural logarithm returns calculated from the composite index of Shanghai Stock Exchange, the Shenzhen Component Index, the Shanghai Stock Fund Index and the Shanghai Stock Bond Index, the vector ARCH with the normal distribution and t distribution, the vector GARCH, Pure diagonal GARCH, BEKK, constant conditional GARCH, principal component GARCH and EWMA models. Based on these models, the VaR is calculated, and then the result of BEKK-t model is used to calculate the VaR of Chinese financial market Value at Risk (VaR) the best results.