Semiparametric GMM Estimation of Spatial Autoregressive Models

We propose semiparametric GMM estimation of semiparametric spatial autoregressive (SAR) models under weak moment conditions. In comparison with the quasi-maximum-likelihood-based semiparametric estimator of Su and Jin (2010), we allow for both heteroscedasticity and spatial dependence in the error t...

Full description

Saved in:
Bibliographic Details
Main Author: SU, Liangjun
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2012
Subjects:
Online Access:https://ink.library.smu.edu.sg/soe_research/1423
https://ink.library.smu.edu.sg/context/soe_research/article/2422/viewcontent/Semi_Parametric_GMM_2013.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
Description
Summary:We propose semiparametric GMM estimation of semiparametric spatial autoregressive (SAR) models under weak moment conditions. In comparison with the quasi-maximum-likelihood-based semiparametric estimator of Su and Jin (2010), we allow for both heteroscedasticity and spatial dependence in the error terms. We derive the limiting distributions of our estimators for both the parametric and nonparametric components in the model and demonstrate the estimator of the parametric component has the usual -asymptotics. When the error term also follows an SAR process, we propose an estimator for the parameter in the SAR error process and derive the joint asymptotic distribution for both spatial parameters. Consistent estimates for the asymptotic variance-covariance matrices of both the parametric and nonparametric components are provided. Monte Carlo simulations indicate that our estimators perform well in finite samples.