Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Shawn Tan <shawnt@student.unimelb.edu.au> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: NLSUR with unbalanced panel |
Date | Fri, 30 Sep 2011 17:38:51 +1000 |
Hi Statalist, I am looking at bilateral trade flows between a group of countries. It is an unbalanced: each exporter (i) does not have the same number of observations (i.e. importers). The regression looks like this: Y(i) = a(i)*p X + a(i) Z There is a exporter-specific parameter, a(i), and a common parameter, p. I want to run a SUR by exporter and impose a constraint that the ratio of the coefficient (= p) is equal across all countries. Initially, I thought I could use the suregub command explained by Baum in "Introduction to Stata Programming" but I don't know how to impose my constraints on the estimation. Then I thought I could use nlsur because it can use the substitutable expressions but nlsur cannot be done for unbalanced panel. I would appreciate any help in either how to impose the constraints with the suregub command or using the nlsur with an unbalanced panel. Thank you. Shawn * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/