A NOTE ON SPURIOUS REGRESSION IN PANELS WITH CROSS-SECTION DEPENDENCE

This paper analyses regression of two independent stationary panels with cross-sectional dependence. It is shown that the pooling least squares (PLS) estimator converges to zero in probability while the individual OLS estimator converges to a random variable. However, the PLS-based and the OLS-based t-statistics diverge, so the null hypothesis of no correlation tends to be spuriously rejected.


Issue Date:
2003
Publication Type:
Working or Discussion Paper
DOI and Other Identifiers:
Record Identifier:
https://ageconsearch.umn.edu/record/23712
PURL Identifier:
http://purl.umn.edu/23712
Total Pages:
12
Series Statement:
Discussion Paper 03.04




 Record created 2017-04-01, last modified 2020-10-28

Fulltext:
Download fulltext
PDF

Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)