响应变量缺失下非线性回归模型的加权半参数估计

Weighted Semiparametric Estimation for Nonlinear Regression Models with Missing Responses at Random

基于逆概率加权方法研究了响应变量缺失下非线性回归模型的参数估计问题,提出了一种利用广义部分线性单指标模型对选择概率建模的加权半参数估计方法.从理论上证明了所得估计量具有渐近正态性,并通过数据模拟分析研究了所提方法在有限样本下的表现.

In this paper, we study the nonlinear regression model based on the inverse probability weighted method, when some response variables are missing at random. We propose a weighted semiparametric estimation method, to model the selection probability by a generalized partially linear single-index model. The resulting estimator is shown to be asymptotic normal under some conditions. Numerical studies based on simulation are used to illustrate the finite sample performance of the proposed procedure.