二维非对称copula函数在干旱特性联合概率中的应用  

Application of two dimensional asymmetric copula function in combined probability of drought characteristics

作　　者：李亦凡[1] 李订芳[1] 

机构地区：[1]武汉大学数学与统计学院,湖北武汉430072

出　　处：《水资源与水工程学报》2016年第5期236-240,共5页Journal of Water Resources and Water Engineering

摘　　要：随着全球气候变化和人类活动的影响,干旱特性分析正成为研究热点。在实际应用中,二维变量常表现出非对称性。基于Liebscher提出的非对称copula函数构造方法构造了二维非对称copula函数。以宜昌水文站日径流量数据为例,运用二维非对称copula函数对干旱历时和干旱峰值的联合概率分布进行分析。证明了构造二维非对称copula时,相关系数的变化规律定理,该定理可以作为copula函数选取的参考准则。采用Rosenblatt变换的Bootstrap法进行copula拟合度检验,检验结果表明二维非对称copula函数可以描述干旱特性的概率分布。With the change of global climate and the influence of human action,the analysis of drought characteristics has become a research hotspot. Because of the asymmetry of 2D variable in practical application,based on the asymmetric copula function proposed by Liebscher,the paper built 2D asymmetric copula function. Taking the daily flood data collected in Yichang hydrological station for example,the paper analyzed the Joint probability distribution of drought duration and peak value by use of 2D asymmetric copula function. It is proved that the law of variation of the correlation coefficient when a two-dimensional asymmetric copula is constructed,which can be used as a reference criterion for the selection of copula function. Bootstrap method of Rosenblatt transform is used to test the fitting degree of copula. Results show that the copula function can describe the probability distribution of drought characteristics.

关 键 词：非对称copula函数 相关系数 干旱特性 联合概率 干旱历时 干旱峰值 

分 类 号：P338.6[天文地球—水文科学]