TY - JOUR
T1 - Some robust approaches to testing and estimation in spatial econometrics
AU - Anselin, Luc
N1 - Funding Information:
*The research of which this paper is an outgrowth was supported in part by Grant SES-8600465 from the National Science Foundation. An earlier version was presented at the 84th Annual Meeting of the Association of American Geographers, Phoenix, AZ, April 6-10, 1988. Helpful comments from an anonymous referee aided in clarifying the writing.
PY - 1990/9
Y1 - 1990/9
N2 - Some robust approaches are outlined that form a basis for a more realistic statistical inference in spatial econometric models. Three specific issues are addressed: significance tests on coefficients in the spatial expansion method that are robust to the presence of heteroskedasticity of unknown form; heteroskedasticity-robust specification tests for spatial dependence; and boot-strap estimation in spatial autoregressive models. The techniques are presented in formal terms and their application to spatial analysis is illustrated in a number of simple empirical examples.
AB - Some robust approaches are outlined that form a basis for a more realistic statistical inference in spatial econometric models. Three specific issues are addressed: significance tests on coefficients in the spatial expansion method that are robust to the presence of heteroskedasticity of unknown form; heteroskedasticity-robust specification tests for spatial dependence; and boot-strap estimation in spatial autoregressive models. The techniques are presented in formal terms and their application to spatial analysis is illustrated in a number of simple empirical examples.
UR - http://www.scopus.com/inward/record.url?scp=0025620895&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0025620895&partnerID=8YFLogxK
U2 - 10.1016/0166-0462(90)90001-J
DO - 10.1016/0166-0462(90)90001-J
M3 - Article
AN - SCOPUS:0025620895
SN - 0166-0462
VL - 20
SP - 141
EP - 163
JO - Regional Science and Urban Economics
JF - Regional Science and Urban Economics
IS - 2
ER -