摘要
研究了协变量维数趋于无穷的复合次序Logisti回归纵向数据模型.首先在响应变量为k个有序"状态"之一时,给出了该模型下的广义估计方程,然后给出了该广义估计方程估计的渐近存在性,相合性以及渐近正态性定理,并在较弱的条件下给出了定理的证明过程,证明了该模型的可用性以及结果的稳定性,推广了文献中的相关结果.
In this paper,we study the compound ordinal logit regression model with diverging number of covariates.First,we propose the GEE of this modle on condation that the response variable denote one of status.Then we bring up the asymptotic existence,consistency and asymptotic normality theorem and under some mild conditons,we provide the evidentiary process.The practicability and stabilization of this model is proved.We extend the relevant results in the literature.
引文
[1]Fahrmeir L,Tutz G.Multivariate Statistcal Modelling Based on Generalized Linear Models Equation Concerning[M].New York:Springer-Verlag,1994.
[2]Liang K Y,Zeger S L.Longitudinal data analysis using generalized linear models[J].Biometrika,1986,73:13-22.
[3]赵目,陈柏成,周勇.纵向数据下广义估计方程估计[J].数学学报:中文版,2012,55(1):1-16.
[4]孙海云.Logit模型参数估计方法的研究[D].浙江:浙江大学,2016.
[5]Wang L.GEE analysis of clustered binary data with diverging number of covariates[J].Ann.Statist.,2011,39:389-417.
[6]Ortega J M,Rheinboldt W C.Iterative Solution of Nonlinear Equations in Several Variables[M].San Diego:Academic Press,1970.