摘要
针对巨灾损失具有尖峰厚尾的特征,采用极值理论中的POT模型拟合巨灾损失变量的边缘分布,利用二元阿基米德Copula函数来刻画巨灾损失变量之间的相关结构,构建了Copula-POT模型来估算巨灾损失的条件VaR值。结合全球洪水事件的损失数据进行实证分析和比较,结果表明:损失变量的厚尾性与相关关系的刻画是估计条件VaR的关键,且Copula-POT模型能够提供准确的巨灾损失条件VaR估计。本研究能为巨灾风险评估及其应对提供一定的参考。
Aiming at the fat tail characteristics of catastrophe losses,this paper applies the POT model to describe the thick tail characteristics of the catastrophe loss variable,and the Archimedean Copula model is used to describe the correlation between the catastrophe loss variables,constructing a Copula-POT model to estimate the value of conditional VaR.An empirical analysis and comparison are carried out based on global floods data and the results show that the thick tail characteristics and the characterization of multivariate correlation are the key to accurately estimate the conditional quantile.Furthermore,the Copula-POT model can provide accurate estimates of the conditional quantile.Thus,the research in this paper can provide a reference for catastrophe risk assessment and risk response.
作者
巢文
钱晓涛
CHAO Wen;QIAN Xiao-tao(School of Management,Fujian University of Technology,Fuzhou 350118,Fujian,China;Department of Basic Teaching and Research,Yango University,Fuzhou 350015,Fujian,China)
出处
《华南理工大学学报（社会科学版）》
2020年第4期88-97,共10页
Journal of South China University of Technology(Social Science Edition)
基金
国家自然科学基金项目(11871152)
福建省自然科学基金项目(2017J01794)
福建省中青年教师教育科研项目(JAS19199)。
