空间回归模型选择的反思

发布时间：2018-02-28 16:37

本文关键词： 空间计量经济学模型空间滞后模型空间误差模型拉格朗日乘子检验　出处：《统计与信息论坛》2016年10期 　论文类型：期刊论文

【摘要】：空间计量经济学存在两种最基本的模型:空间滞后模型和空间误差模型,这里旨在重新思考和探讨这两种空间回归模型的选择,结论为:Moran’s I指数可以用来判断回归模型后的残差是否存在空间依赖性;在实证分析中,采用拉格朗日乘子检验判断两种模型优劣是最常见的做法。然而,该检验仅仅是基于统计推断而忽略了理论基础,因此,可能导致选择错误的模型;在实证分析中,空间误差模型经常被选择性遗忘,而该模型的适用性较空间滞后模型更为广泛;实证分析大多缺乏空间回归模型设定的探讨,Anselin提出三个统计量,并且,如果模型设定正确,应该遵从Wald统计量Log likelihood统计量LM统计量的排列顺序。
[Abstract]:There are two basic models in spatial econometrics: spatial lag model and spatial error model. The conclusion is that the ratio Moranzi I index can be used to judge the spatial dependence of the residual error after regression model, and the Lagrange multiplier test is the most common method to judge the advantages and disadvantages of the two models in the empirical analysis. The test ignores the theoretical basis only on the basis of statistical inference, and therefore may lead to the selection of the wrong model; in empirical analysis, spatial error models are often selectively forgotten. However, the applicability of the model is more extensive than that of the spatial lag model; the empirical analysis mostly lacks the discussion of spatial regression model setting. Anselin proposes three statistics, and if the model is set correctly, Wald statistics should be followed in the order in which Log likelihood statistics LM statistics are arranged.
【作者单位】：浙江财经大学经济学院;
【分类号】：F224.0

【相似文献】