Publication Type
Working Paper
Version
publishedVersion
Publication Date
9-2014
Abstract
In the presence of heteroskedasticity, Lin and Lee (2010) show that the quasi maximum likelihood (QML) estimators of spatial autoregressive models (SAR) can be inconsistent as a ‘necessary’ condition for consistency can be violated, and thus propose robust GMM estimators for the model. In this paper, we first show that this condition may hold in many practical situations and when it does the regular QML estimators can be consistent.In cases where this condition is violated, we propose a modified QML estimation method robust against heteroskedasticity of unknown form. In both cases, asymptotic distributions of the estimators are derived, and methods for estimating robust variances are given, leading to robust inferences for the model. Extensive Monte Carlo results show that the modified QML estimator outperforms the GMM estimators, and the regular QML estimator even when it is consistent. The proposed robust inference methods can also be easily applied.
Keywords
Spatial dependence, Unknown heteroskedasticity, Nonnormality, Modified QML estimator, Robust standard error
Discipline
Econometrics
Research Areas
Econometrics
First Page
1
Last Page
35
Publisher
SMU Economics and Statistics Working Paper Series, No. 14-2014
City or Country
Singapore
Citation
LIU, Shew Fan and YANG, Zhenlin.
Modified QML Estimation of Spatial Autoregressive Models with Unknown Heteroskedasticity and Nonnormality. (2014). 1-35.
Available at: https://ink.library.smu.edu.sg/soe_research/1598
Copyright Owner and License
Authors
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.
Comments
Published in Regional Science and Urban Economics https://doi.org/10.1016/j.regsciurbeco.2015.02.003