论文部分内容阅读
在双层嵌套模拟中,通常外层模拟生成场景,内层样本估计在给定场景下的条件期望方差.所引入场景效应和观察误差随机变量,使条件期望的方差估计量是无偏的,并且讨论了计算量一定的情况下使估计量的方差达到最小的最优内层样本容量.然后,将该方法运用到金融风险度量中,分析了收益率均值和标准差对VaR的影响.最后,基于双层嵌套模拟提出一种估计VaR的新方法,该方法既能有效地处理非线性非正态的情形,又在一定程度上解决了参数选择的问题.实证研究结果通过了频率失败检验,说明该方法的合理有效性.
In a two-level nested simulation, the outer layer is usually simulated to generate a scene, and the inner sample to estimate the conditional expected variance in a given scene. The introduced scenario effects and observed error random variables make the variance estimate of conditional expectation unbiased , And discusses the optimal inner sample size that minimizes the variance of the estimator with a certain amount of computation.Then, this method is applied to the financial risk measure to analyze the impact of the mean and standard deviation of return on VaR. Finally, a new method of estimating VaR is proposed based on two-level nested simulation, which not only can effectively deal with nonlinear non-normal situation, but also solves the problem of parameter selection to a certain extent.Experimental results pass the frequency Failure to test, indicating that the method is reasonable and effective.