多维相关随机变量的抽样方法  

Sampling Method for Multidimensional Correlated Random Variables

作　　者：魏艳华[1] 王丙参[1,2] Wei Yanhua;Wang Bingcan(School of Mathematics and Statistics,Tianshui Normal University,Tianshui Gansu 741001,China;School of Statistics,Capital University of Economics and Business,Beijing 100070,China)

机构地区：[1]天水师范学院数学与统计学院,甘肃天水741001 [2]首都经济贸易大学统计学院,北京100070

出　　处：《统计与决策》2020年第17期5-8,共4页Statistics & Decision

基　　金：国家自然科学基金资助项目(11665019);天水师范学院中青年教师科研资助项目(ZD2017-01)。

摘　　要：文章首先利用乔里斯基分解方法生成相关正态随机变量的抽样序列,然后利用两步变换法生成非正态随机变量的抽样序列,给出了非线性变换的确定方法,在确定中间正态变量的相关系数矩阵时,建议优先使用解析方法,其次考虑数值方法,探讨了插值拟合数值计算方法的优缺点与两步变换法的失效情况,并利用相关指数随机数生成相关威布尔分布随机数。最后结合实例验证了理论分析。This paper first use the Jorisky decomposition method to generate the sampling sequence of correlated normal random variables,and then employs the two-step transformation method to generate a sampling sequence of non-normal random variables,presenting the method of determining the nonlinear transformation.When determining the correlation coefficient matrix of the intermediate normal variable,the paper recommends using the analytical method first,and then considering the numerical method.The paper also discusses the merits and demerits of interpolation fitting numerical calculation method and the failure of two-step transformation method,generating the related random number of Weibull distribution by use of the correlation index random number.Finally,an example is given to verify the theoretical analysis.

关 键 词：相关随机变量 随机数 变换算法 数值计算 

分 类 号：O212[理学—概率论与数理统计]