横截面独立的长相依面板数据的公共均值变点  

A common break in means for long-range dependent panel data under cross-sectional independence

作　　者：习代青 傅承德 庞天晓[3] XI Dai-qing;FUH Cheng-Der;PANG Tian-xiao(School of Statistics and Mathematics,Zhongnan University of Economics and Law,Wuhan 430073,China;Graduate Institute of Statistics,Central University,Taoyuan(Taiwan)320317,China;School of Mathematical Sciences,Zhejiang University,Hangzhou 310058,China)

机构地区：[1]中南财经政法大学统计与数学学院,湖北武汉430073 [2]中央大学统计研究所,中国台湾桃园320317 [3]浙江大学数学科学学院,浙江杭州310058

出　　处：《高校应用数学学报（A辑）》2024年第3期253-272,共20页Applied Mathematics A Journal of Chinese Universities(Ser.A)

基　　金：国家社会科学基金(21BTJ067)。

摘　　要：研究横截面独立的长相依面板数据的公共均值变点的估计问题.文中分别在强变点信号,中等变点信号和弱变点信号这三种情形下研究了公共变点的估计量,建立了估计量的渐近性质,包括相合性,收敛速度和极限分布.理论结果表明在变点信号和长相依之间存在着一种权衡关系.具体来说,在强变点信号情形下,长相依不影响估计量的渐近性质;在中等变点信号情形下,长相依不影响估计量的收敛速度,但影响估计量的极限分布;在弱变点信号情形下,长相依既影响估计量的收敛速度,也影响估计量的极限分布.Monte Carlo模拟评估了估计量在有限样本情形下的表现,并支持文中的理论结果.This paper focuses on estimating a common break point in means for long-range depen-dent panel data under cross-sectional independence.The common break-point estimator is examined under three scenarios:strong,moderate and weak break signals.Asymptotic properties,including consistency,rate of convergence and limiting distribution,of the estimator are established.The theo-retical results reveal that there is a trade-off between the break signal and long-range dependence.To be more precise,the long-range dependence has no ability to influence the asymptotic behaviors of the estimator if the break signal is strong,it does not influence the rate of convergence but has an impact on the limiting distribution of the estimator when the break signal is moderate,and it influences both the rate of convergence and the limiting distribution of the estimator when the break signal is weak.Monte Carlo simulations are conducted to assess thefinite-sample performance of the estimator,and the theoretical results are supported by the simulation results.

关 键 词：公共变点 极限分布 长相依 面板数据 

分 类 号：O212.1[理学—概率论与数理统计]