空间自回归模型误差项的空间相关性检验  

Spatial Correlation Test of Error Terms in Spatial Autoregressive Models

作　　者：惠姣姣 Hui Jiaojiao(School of Mathematics and Information Science,Baoji University of Arts and Sciences,Baoji Shaanxi 721013,China)

机构地区：[1]宝鸡文理学院数学与信息科学学院,陕西宝鸡721013

出　　处：《统计与决策》2025年第7期47-52,共6页Statistics & Decision

基　　金：陕西省教育厅专项科学研究计划项目(22JK0250)。

摘　　要：近年来,随着现代技术的蓬勃发展,空间相关性问题的重要性日益凸显,由于空间数据之间大多存在相依性,因此对数据进行空间相关性检验就显得尤为重要。基于这一现实背景,文章首先利用两阶段最小二乘法对空间自回归模型的参数进行估计,并利用三阶矩χ2逼近方法对空间自回归模型的误差项进行空间相关性检验;其次,进行Monte-Carlo模拟分析,结果显示,空间自回归模型的参数估计和误差项空间相关性检验效果都很好;最后,对我国31个省份的居民福利水平及其影响因素进行实证分析,验证了空间自回归模型的参数估计和误差项的空间相关性检验具有实际应用意义。In recent years,with the vigorous development of modern technology,the importance of spatial correlation has become increasingly prominent.Due to the interdependence between spatial data,it appears particularly important to conduct spatial correlation test on data.Based on the reality,this paper first uses the two-stage least squares method to estimate the parameters of the spatial autoregressive model,and uses the third-order moment χ^(2) approximation method to perform spatial correlation test on the error terms of the spatial autoregressive model,and then performs the Monte Carlo simulation analysis.The results show that the parameter estimation and spatial correlation test of error terms of the spatial autoregressive model are very effective.Finally,the paper makes an empirical analysis on the residents’welfare level and its influencing factors in 31 provinces of China to verify that the parameter estimation of the spatial autoregressive model and the spatial correlation test of the error term have practical application significance.

关 键 词：空间自回归模型 空间相关性检验 两阶段最小二乘法 三阶矩χ2逼近方法 

分 类 号：O212[理学—概率论与数理统计]