Random Threshold Driven Tail Dependence Measures with Application to Precipitation Data Analysis
报告人：张正军
报告地点：数学与统计学院403室
报告时间：2013年12月25日星期三16：00:17：00
邀请人：
报告摘要：
Identification of tail dependence among observations is important and challenging, but remains an open problem, due to the fact that tail dependence is primarily captured by values above thresholds. This paper introduces a class of tail quotient correlation coefficients (TQCC) which allows the underlying threshold values to be random. An approximation theorem of conditional tail probabilities is established. The limiting distribution of TQCC under the null hypothesis of tail independence is derived. Test statistics for tail independence are constructed and shown to be consistent under the alternative hypothesis of tail dependence. We apply TQCC to investigate tail dependencies of daily precipitation in continental US. Our results indicate nonstationarity, spatial clusters, and tail dependence in the data.
主讲人简介：
Identification of tail dependence among observations is important and challenging, but remains an open problem, due to the fact that tail dependence is primarily captured by values above thresholds. This paper introduces a class of tail quotient correlation coefficients (TQCC) which allows the underlying threshold values to be random. An approximation theorem of conditional tail probabilities is established. The limiting distribution of TQCC under the null hypothesis of tail independence is derived. Test statistics for tail independence are constructed and shown to be consistent under the alternative hypothesis of tail dependence. We apply TQCC to investigate tail dependencies of daily precipitation in continental US. Our results indicate nonstationarity, spatial clusters, and tail dependence in the data.
举办单位：数学与统计学院 阅读次数：次
学校主页