Cover Image

Spatial lag test with equal weights

This note shows that for a spatial regression with equal weights, the LM test is always equal to N / 2(N - 1), where N is the sample size. This means that this test statistics is a function of N and not a function of the spatial parameter [rho]. In fact, this test statistic tends to one half for N tending to infinity. The null hypothesis of no spatial correlation is never rejected no matter what [rho] is.
Year of publication:
2009
Authors: Baltagi, Badi H. ; Liu, Long
Published in:
Economics Letters. - Elsevier, ISSN 0165-1765. - Vol. 104.2009, 2, p. 81-82
Publisher: Elsevier
Keywords: Spatial error correlation Equal weights Lagrange multiplier
Online Resource
Check full text access |
Check Google Scholar
In libraries world-wide (WorldCat)
In German libraries (KVK)
subito order
I need help
Type of publication: Article
Source:
Persistent link: https://www.econbiz.de/10005023468
Similar items by person