本文考虑如下纵向数据半参数回归模型:yij=xij′β+g(tij)+eij。基于最小二乘法和一般的非参数权函数方法给出了模型中参数β,回归函数g(·)和误差方差σ^2的估计,并在适当条件下证明了估计量的强相合性。
We consider the following semiparametric regression model for longitudinal data: yij=xij′β+g(tij)+eij. The estimators of β, g(·) and σ^2 are obtained by using the least squares and usual nonparametric weight function method. We prove their strong consistency under suitable conditions.