半参数空间自回归变系数模型的统计推断  

Statistical Inference of Semi-parametric Spatial Autoregressive Variable Coefficient Model

作　　者：陈凤 刘嘉慧 Chen Feng;Liu Jiahui(School of Mathematics and Statistics,Chongqing Jiaotong University,Chongqing 400074,China;School of Management,Xi’an Jiaotong University,Xi’an 710049,China)

机构地区：[1]重庆交通大学数学与统计学院,重庆400074 [2]西安交通大学管理学院,西安710049

出　　处：《统计与决策》2024年第4期17-22,共6页Statistics & Decision

基　　金：国家社会科学基金重点项目(20AGL004)。

摘　　要：半参数空间自回归变系数模型因同时考虑了因变量的空间自相关性和回归关系的空间异质性而具有广阔的应用前景。文章针对此模型,基于轮廓拟极大似然估计,提出了常值系数和局部系数的显著性检验方法,以辨识模型中的零值系数,从而更深入地理解回归关系的变化特征;同时,也给出了局部检验可能涉及的多重检验的解决方法。模拟实验验证了所提检验方法的有效性,而基于美国波士顿房屋价格数据的实证分析验证了检验方法的应用效果。The semi-parametric spatial autoregressive variable coefficient model has a broad application prospect because it takes into account both the spatial autocorrelation of dependent variables and the spatial heterogeneity of regression relations.Based on contour quasi-maximum likelihood estimation,this paper proposes a significance test method of constant coefficient and local coefficient to identify the zero-value coefficient in the model,so as to understand the change characteristics of regression relations more deeply.At the same time,is also given the solution of multiple tests which may be involved in local tests.Simulation experiments verify the validity of the proposed test method,and the empirical analysis based on the house price data of Boston in USA verifies the application effect of the test method.

关 键 词：空间非平稳性 空间自相关性 统计推断 多重检验 

分 类 号：F224[经济管理—国民经济] O212.7[理学—概率论与数理统计]