关键词:K-M估计 均方误差 Wiener过程
作者简介:欧阳资生湖南商学院数学系,长沙410205;周光明长沙铁道学院数学系,长沙410075
引言
设X[,1],X[,2],…,X[,n]为一列非负的i.i.d.随机变量序列,分布函数为F(x),密度函数为f(x)=F′(x)。现在我们假定{X[,i],i=1,2,…}被另一族随机变量序列{Y[,i],i=1,2,…}所截断,且{Y[,i],i=1,2,…}也是非负i.i.d.的,其分布函数为G(t),且设其连续,我们仅能观察到样本数据:
研究了用f[,n](x)作为f(x)的估计时的均方误差,得出了与Zhang[5]相类似的结果。
在(2)中,K(t)是一适当的核函数,ρ[,n]=ρ[,n,k](x)是在Z[,1],Z[,2],…,Z[,n]中x到它的第k个最近邻的距离,k=k[,n]是一列正整数,满足
参考文献:
1Burke,M.D.,,S.,Horváth,L.,A correction to and improvement of strong approximations of some biometrics estimates under random censorship,Probab.Theory Related Fields,79(1988),51-57.
2Kaplan.E.& Meier,P.,Nonparametric estimation from incomplete observation,J.A.S.A.,53(1958),457-481.
3Stefan,S.R.,The law of the iterated logarithm for the multivariate nearest neighbor density estimators,J.Multi.Anal.,53(1995),159-157.
4,H.,Local convergence of empirical measures in the random censorship situation with applications to density and rate estimators,Ann.Stat.,14(1986),1240-1245.
5Zhang B,A note on the integrated square errors of kernel density estimators under random censorship,Stochastic Process Appl.,75(1998),235-248.