纵向数据下部分非线性模型的广义经验似然推断

摘要

In this paper, we study the construction of confidence region for unknown parameter in partially nonlinear models with longitudinal data. By empirical likelihood method, the generalized empirical log-likelihood ratio for parameter in nonlinear function is proposed and shown to be asymptotically chi-square distribution. At the same time, the maximum empirical likelihood estimator of the parameter in nonlinear function is obtained and asymptotic normality is proved.%本文研究了纵向数据下部分非线性模型中未知参数的置信域的构造。利用经验似然方法，构造了非线性函数中未知参数的广义对数经验似然比统计量，证明了其渐近于卡方分布。同时，得到了未知参数的最大经验似然估计，并证明了其渐近正态性。