摘要
设{εi,1≤i≤n}为ND随机误差序列,利用ND序列的Bernstein不等式,在非参数回归模型Yi=g(xi)+εi(1≤i≤n)下,研究未知函数g(x)加权核估计gn(x)=∑n i=1 Yi(xi-xi-1)/hn K((x-xi)/hn)的强相合速度,从而将加权回归函数估计的相合性推广到ND样本.
Let {εi,1≤i≤n} be negatively dependent and random error sequence.With the help of the nonparametric regression model Yi=g（xi）＋εi（1≤i≤n）,we discussed the rate of the strong consistency of the nonparametric regression weighted function estimator gn（x）=∑n i=1Yi（xi-xi-1）/hn K（（x-xi）/hn） for negatively dependent samples using Bernstein inequality.The consistency was extended to the case of negatively dependent samples.
出处
《吉林大学学报（理学版）》
CAS
CSCD
北大核心
2013年第2期237-240,共4页
Journal of Jilin University:Science Edition
基金
国家自然科学基金(批准号:11061012)
广西自然科学基金(批准号:2012GXNSFAA053010)
关键词
ND序列
非参数回归
加权核估计
强相合速度
negatively dependent sequence
nonparametric regression
weighted kernel estimator
rate of strong consistency
