Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/268377 
Authors: 
Year of Publication: 
2023
Series/Report no.: 
IHS Working Paper No. 44
Publisher: 
Institut für Höhere Studien - Institute for Advanced Studies (IHS), Vienna
Abstract: 
We consider fully modified least squares estimation for systems of cointegrating polynomial regressions, i. e., systems of regressions that include deterministic variables, integrated processes and their powers as regressors. The errors are allowed to be correlated across equations, over time and with the regressors. Whilst, of course, fully modified OLS and GLS estimation coincide - for any regular weighting matrix - without restrictions on the parameters and with the same regressors in all equations, this equivalence breaks down, in general, in case of parameter restrictions and/or different regressors across equations. Consequently, we discuss in detail restricted fully modified GLS estimators and inference based upon them.
Subjects: 
Fully Modified Estimation
Cointegrating Polynomial Regression
Generalized Least Squares
Hypothesis Testing
JEL: 
C12
C13
Q20
Creative Commons License: 
cc-by Logo
Document Type: 
Working Paper

Files in This Item:
File
Size
510.91 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.