Function stslshac {sphet}: heteroskedasticity and autocorrelation consistent (HAC) estimator
Dear Roger, Thank you for your reply. Finally I got reply from Harry Kelejian, the author of the paper referred to in R help for the function stslshac and also the author of the paper you are referring to. His response was: 'The HAC estimator allows for spatial correlation , as well as heteroskedasticity. There is no need to test for spatial correlation or heteroskedasticity. when using the HAC.' Which answered my question and made this issue clear for me. Regarding your reference, the answer may be implied in the paper, however I would need to spend quite long time studying it to understand it and unfortunately I am quite busy at the moment. Thank you for willingness to help anyway. Monika
On 21 September 2015 at 17:04, Roger Bivand <Roger.Bivand at nhh.no> wrote:
On Sat, 19 Sep 2015, monika nov wrote:
Dear R-users, I have quite basic question for econometricians, however I would like to be sure in this. If I use a HAC estimator of the variance-covariance (VC) matrix for a spatial econometric model, do I still need to test the residuals for spatial autocorrelation and heteroscedasticity? (in particular I am using function stslshac available in package sphet. The estimator is based on Kelejian, H.H. and Prucha, I.R. (2007) HAC estimation in a spatial framework, Journal of Econometrics, 140, pages 131?154). What if the residuals from model estimated by stslshac are spatially autocorrelated and (or) heteroscedastic? Can I still use this estimator with HAC estimate of VC matrix or shall I go for different estimator or specification? Do the estimates have required properties (are they unbiased, consistent, efficient)? I would be grateful for any reaction.
Does this reference throw any light on the question? I'm not aware of
implementations:
@article{
year={2015},
issn={1864-4031},
journal={Letters in Spatial and Resource Sciences},
doi={10.1007/s12076-015-0146-2},
title={Critical issues in spatial models: error term specifications,
additional endogenous variables, pre-testing, and Bayesian analysis},
url={http://dx.doi.org/10.1007/s12076-015-0146-2},
publisher={Springer Berlin Heidelberg},
keywords={Specifications of spatial models; Additional endogenous variables;
Pre-testing; Bayesian analysis; C01; C12; C13},
author={Kelejian, HarryH.},
pages={1-24},
language={English}
}
Roger
_______________________________________________ R-sig-Geo mailing list R-sig-Geo at r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-geo
-- Roger Bivand Department of Economics, Norwegian School of Economics, Helleveien 30, N-5045 Bergen, Norway. voice: +47 55 95 93 55; fax +47 55 95 91 00 e-mail: Roger.Bivand at nhh.no