The Econometric Society An International Society for the Advancement of Economic Theory in its Relation to Statistics and Mathematics
Home Contacts
Econometrica

New Journals

Econometrica
Editorial Board
Journal News

Monograph Series

October 1969 - Volume 37 Issue 4 Page 651 - 659


p.651


The Consistency and Efficiency of Generalized Least Squares in Simultaneous Equation Systems with Autocorrelated Errors

M. R. Wickens

Abstract

Several existing single equation estimators, designed for the case where the structural disturbances are generated by a first order autoregressive process, are generalized for situations where the autoregressive process is of higher order than one. These generalized estimators are shown to be consistent, both when the parameters of the generating process are known and unknown; and their asymptotic relative efficiencies are compared. Three full information estimators are also examined. They are shown to be consistent with asymptotic efficiencies ranked like their single equation counterparts. Among the results derived are the consistency of 2SLS and 3SLS in this situation and the overall relative efficiencies of the "Theil-Fisher" type estimators compared with the 2SLS-3SLS and "Madansky" type estimators.

Full content Login                                    

Note: to view the fulltext of the article, please login first and then click the "full content" button. If you are based at a subscribing Institution or Library or if you have a separate access to JSTOR/Wiley Online Library please click on the "Institutional access" button.
Prev | All Articles | Next
Go to top
Membership



Email me my password
Join/Renew
Change your address
Register for password
Require login:
Amend your profile
E-mail Alerting
The Society
About the Society
Society News
Society Reports
Officers
Fellows
Members
Regions
Meetings
Future Meetings
Past Meetings
Meeting Announcements
Google
web this site
   
Wiley-Blackwell
Site created and maintained by Wiley-Blackwell.
Comments? Contact customsiteshelp@wiley.com
To view our Privacy Policy, please click here.