非线性模型误差方差估计的Bootstrap逼近和分布的收敛速度  

Convergence Rates of the Distributions and the Bootstrap Approximations of the Error Variance Estimates in Nonlinear Models

作　　者：黄养新[1] 

机构地区：[1]武汉工业大学,武汉430070

出　　处：《应用数学》1994年第1期11-17,共7页Mathematica Applicata

摘　　要：本文对非线性模型误差方差的估计基于Jackknife虚拟值的Bootstrap方法建立了Bootstrap逼近,证明了逼近的相合性定理,得到了逼近的速度是o(n^(-1/2))。进一步,本文证明了误差方差估计的分布以理想的最佳速度o(n^(-1/2))收敛于正态分布的结论。In this paper,the Bootstrap approximations based on the Jackknife pseudovalues is applied to the estimation of error variance in nonlinear models. The large sample properties of the approximations is considered, it is proved that the approximation is consistent and convergence rate is o( n-1/2). Furthermore,the convergence rate of the distribution of error variance estimates is presented.

关 键 词：非线性模型 误差 方差 估计 

分 类 号：O212.1[理学—概率论与数理统计]