REVISITING ERROR AUTOCORRELATION CORRECTION: COMMON FACTOR RESTRICTIONS AND GRANGER CAUSALITY

This paper demonstrates that linear regression models with an AR(1) error structure implicitly assume that y{t} does not Granger cause any of the exogenous variables in X{t}. An indirect test of the common factor restrictions based on this Granger non-causality is proposed and shown to outperform existing tests.


Issue Date:
2004
Publication Type:
Conference Paper/ Presentation
DOI and Other Identifiers:
Record Identifier:
https://ageconsearch.umn.edu/record/20176
PURL Identifier:
http://purl.umn.edu/20176
Total Pages:
24
Series Statement:
Selected Paper




 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)