摘要
设(X1,Y1),(X2,Y2),…,(Xn,Yn)为取值于Rd×R1上的一组样本,在独立同分布(i.i.d)样本下,构造了截尾数据时回归函数基于分割估计及改良基于分割估计,并获得了其强相合性;在同分布φ-mixing相依样本下,获得了回归函数基于分割估计及改良基于分割估计的强相合性及收敛速度。
Let （X1 ,Y1）, （X2,Y2） , （Xn,Yn） be random vectors taking values in R^d×R^1. Then based on the independent identical distribution（i, i. d） samples, partitioning estimation and modified partitioning estimation for regression functions with censored data are constructed ,and their strong consistency is obtained. Also presented are the strong consistency and the rate of convergence of modified partitioning estimation for regression functions under some dependence samples, such as ψ-mixing samples with identical distribution.
出处
《合肥工业大学学报（自然科学版）》
CAS
CSCD
北大核心
2005年第9期1223-1225,共3页
Journal of Hefei University of Technology：Natural Science
基金
合肥工业大学科研发展基金资助项目(041002F)
关键词
回归函数
分割估计
改良分割估计
截尾样本
ψ-mixing样本
强相合性
regression function
partitioning estimation
modified partitioning estimation
censored data
ψ-mixing sample
strong consistency
