混合样本下非参数核回归估计的渐近性质  

Asymptotic Properties of Nonparametric Kernel Regression Estimator for -Mixing Samples

作　　者：杨秀桃[1,2] 杨善朝 YANG Xiutao;YANG Shanchao(Natural science Teaching Department, Haikou college of Economics, Haikou 571127, China;School of Mathematics and Statistics, Guangxi Normal University, Guilin 5~100~, China)

机构地区：[1]海口经济学院自然科学教学部,海南海口571127 [2]广西师范大学数学与统计学院,广西桂林541004

出　　处：《应用数学》2018年第2期422-428,共7页Mathematica Applicata

基　　金：海南省自然科学基金(117173);国家自然科学基金(11061007)

摘　　要：本文在混合样本下讨论Priestley和CHAO(1972)提出的一类非参数核回归估计的渐近性质,在较弱的条件下证明了该估计的完全收敛性与强相合性.For -mixing samples, we discuss the asymptotic properties of the nonparametric kernel regression estimator proposed by Priestley and Chao（1972）. Under more weaker conditions, its complete convergence and strong consistency are proved.

关 键 词：混合样本 非参数核回归估计 完全收敛性 强相合性 

分 类 号：O212.7[理学—概率论与数理统计]