时空变系数半参数模型的轮廓最小二乘估计

Profile Least-squares Estimation of Time-space Variable Coefficients Semi-parametric Models

就时空变系数半参数模型进行研究,给出了该类模型的轮廓最小二乘估计方法,得到模型中时空变系数及常值系数估计的解析表达式,并进行数值模拟予以验证.结果表明,常系数的估计具有无偏性,响应变量估计值与真实值拟合程度较好,相对误差较小.

In this paper,the time-space variable coefficient semi-parametric model is studied,the profile least squares estimation method for this kind of models is proposed,the analytical expressions of time-space variable coefficient and constant coefficient estimation in the model are obtained and verified by numerical simulation. The results show that the estimation of constant coefficients is unbiased,the estimated value of response variable fits well with the true value,and the relative error is small.