介绍了纵向数据非线性混合效应模型,将有界函数Huber函数引入该模型的对数似然函数,得到模型参数的稳健估计(M估计),并探讨了基于M估计异常点检验的问题,通过血浆药物渗透数据说明了方法是有效的.
This paper focus on nonlinear mixed effects model for longitudinal data. Huber's function was introduced in the log-likelihood function of the model and M-estimations are got through Fisher-scoring method. Then the test for outliers in the model based on M-estimation is also studied. At last, the methods are illustrated by a data example.