带自适应设计的拟似然非线性模型中极大拟似然估计的强相合性  

Strong Consistency of Maximum Quasi-Likelihood Estimator in Quasi-Likelihood Nonlinear Models with Adaptive Designs

作　　者：夏天[1] 王学仁[2] 

机构地区：[1]贵州财经大学数学与统计学院,贵州贵阳550025 [2]云南大学统计系,云南昆明650091

出　　处：《数学的实践与认识》2015年第21期252-258,共7页Mathematics in Practice and Theory

基　　金：国家自然科学基金(11361013;11171293);贵州省科学技术基金((2008)2249)

摘　　要：在带自适应设计的拟似然非线性模型中,在响应变量的矩条件尽可能弱和其它正则条件下,证明了以概率为1,当n充分大时,拟似然方程有一个解βn,它收敛于参数真值β0.In quasi-likelihood nonlinear models with adaptive designs, under the moment condition on responses as weak as possible and other mild regular conditions, the authors prove that with probability one, the quasi-likelihood equation has a solution βn for all large sample size n, which converges to the true regression parameter β0.

关 键 词：带自适应设计的拟似然非线性模型 强相合性 极大拟似然估计 渐近性质 

分 类 号：O212.1[理学—概率论与数理统计]