Please use this identifier to cite or link to this item: http://hdl.handle.net/10419/153668
Authors: 
Jarociński, Marek
Year of Publication: 
2010
Series/Report no.: 
ECB Working Paper 1234
Abstract: 
The number of variables related to long-run economic growth is large compared with the number of countries. Bayesian model averaging is often used to impose parsimony in the cross-country growth regression. The underlying prior is that many of the considered variables need to be excluded from the model. This paper, instead, advocates priors that impose parsimony without excluding variables. The resulting models fit the data better and are more robust to revisions of income data. The positive relationship between measures of trade openness and growth is much stronger than found in the literature.
Subjects: 
Adaptive Ridge Regression
Bayesian model averaging
Economic Growth
measurement error
JEL: 
C20
C52
O40
O47
Document Type: 
Working Paper

Files in This Item:
File
Size





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