Menu Content/Inhalt
Seminars Print
previous year previous month next month next year
See by year See by month See by week See Today Search Jump to month
Marek Jarocinski, ECB Print
Thursday, 01 April 2010, 12:15 - 13:15

Implosing Parsimony in Cross-Country Growth Regressions

Marek Jarocinski, ECB

Abstract: The number of variables potentially related to long-run economic growth is large compared with the number of countries.  A popular approach to imposing parsimony in cross-country growth regressions, without making strong assumptions about the specification is to use Bayesian model averaging.  This approach reflects a prior that many of the considered variables need to be excluded from the model, although we do not know which ones. This paper shows that priors which impose parsimony in the model with all considered variables, without excluding any of them, fit the cross-country data better and imply results which are more robust to income data revisions.  With these priors, a positive relationship between measures of trade openness and growth is much stronger than found in the literature.

Location: S 12.227
Contact: Claude Adan, This e-mail address is being protected from spam bots, you need JavaScript enabled to view it