统计研究 ›› 2018, Vol. 35 ›› Issue (4): 117-128.doi: 10.19343/j.cnki.11-1302/c.2018.04.011

• • 上一篇    

平滑转移空间自回归模型下IV方法参数估计值的一致性研究

郭光远等   

  • 出版日期:2018-04-25 发布日期:2018-04-25

Research on the Consistency of IV Estimator under Smooth Transition Spatial Autoregressive Models

GuoGuangyuan et al.   

  • Online:2018-04-25 Published:2018-04-25

摘要: 本文研究了空间自回归模型的一种非线性形式:平滑转移空间自回归模型。该模型空间项系数与转移函数的形式相关,随转移变量变化而变化,既能刻画个体间的关联性,又能描述空间关联性随某些因素变化而发生的改变。本文在工具变量框架下主要讨论了Logistic平滑转移函数空间自回归模型的一些性质,对Exponential转移函数模型也作了相应比较分析,并给出了一系列的设定、检验、估计等过程的详细步骤。在较宽泛的假设条件下,我们证明了模型参数估计值的一致性,并对其进行了Monte Carlo模拟验证,模拟结果很好的支持了一致性结论。

关键词: 平滑转移, 空间自回归, 工具变量, Logistic转移函数

Abstract: In this paper, we study a nonlinear form of the spatial auto-regression model: Smooth Transition Spatial Autoregressive Model (STSAR). In this model, the space term is nonlinearly related to the transition variables, which can not only describe the correlation between individuals, but also depict the changes of such correlations. Based on instrument variable methods (IV), firstly we discuss some properties of the Logistic STSAR (LSTSAR) and Exponential STSAR (ESTSAR), then make a comparative analysis with such two models, and give a detailed introduction of these models’ specification and estimation. Provided that certain regularity conditions are met, we show that the IV estimators are consistent. Moreover, Monte Carlo simulations also fairly support our propositions.

Key words: Smooth Transition, Spatial Autoregressive, Instrument Variable, Logistic Transition