统计与决策2024，Vol.40Issue(15)：52-58,7.DOI:10.13546/j.cnki.tjyjc.2024.15.009
异方差混合地理加权回归模型的正交投影估计
Orthogonality Projection-based Estimation for Mixed Geographically Weighted Regression Models With Heteroscedasticity
摘要
Abstract
This paper proposes a new estimation method for the mixed geographically weighted regression models with spatial heteroscedasticity of error terms.This method combines the idea of orthogonal projection,local linear estimation,and generalized least squares estimation to separately estimate the constant coefficients,coefficient functions and variance functions in the model.The performance of the proposed method is verified through numerical simulation.The simulation results show that the proposed method performs better than existing reweighted estimation methods.Finally,the practicality of the proposed method is verified through empirical research based on the cultural consumption and influencing factors of urban residents.关键词
混合地理加权回归模型/正交投影估计/空间异方差性/局部线性估计Key words
mixed geographically weighted regression model/orthogonal projection estimation/spatial heteroscedasticity/lo-cal linear estimation分类
数理科学引用本文复制引用
古丽斯坦·库尔班尼牙孜,孟丽君,田茂再..异方差混合地理加权回归模型的正交投影估计[J].统计与决策,2024,40(15):52-58,7.基金项目
新疆维吾尔自治区社会科学基金项目(22VZX021) （22VZX021）
新疆财经大学高层次人才专项项目(2023XGC006) （2023XGC006）
新疆维吾尔自治区普通高校人文社会科学重点研究基地基金项目(XJEDU2023P011) （XJEDU2023P011）
