Some Evidence on the Potential of Theil's Generalized Two Stage Least Squares Estimator

Adolf Buse () and Bakhtiar Moazzami

Empirical Economics, 1991, vol. 16, issue 3, 335-49

Abstract: This paper evaluates the small sample performance of Theil's generalized two stage least squares (G2SLS) estimator for a structural equation with first order autocorrelation. We find that the bias of G2SLS and a variant thereof is always smaller than that of conventional estimators such as Sargan's 2SLS and Fair's method. The Theil estimators also dominate the Fair estimator in MSE and one of them shares first place in the MSE rankings with an asymptotically efficient modified Sargan 2SLS estimator. Because G2SLS uses a much smaller instrument set than the efficient Sargan estimators it provides a useful and practical solution to the degrees of freedom problem.

Date: 1991
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:empeco:v:16:y:1991:i:3:p:335-49

Ordering information: This journal article can be ordered from
http://www.springer. ... rics/journal/181/PS2

Access Statistics for this article

Empirical Economics is currently edited by Robert M. Kunst, Arthur H.O. van Soest, Bertrand Candelon, Subal C. Kumbhakar and Joakim Westerlund

More articles in Empirical Economics from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().