Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | SIRENIA VAZQUEZ BAEZ <sirenia.vazquez@bbva.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: Margins command to calculate elasticities using a heckman model |
Date | Mon, 7 Oct 2013 10:21:52 -0500 |
Dear Statalist I'm new with the -margins- command and I'm using it after running a Heckman model in Stata 13. I'm working with the eyex() option to estimate the elasticity of credit card purchases to credit cards interest rates using panel data. My dependent variable is the amount of credit card purchases, and as explanatory variables I have the interest rate (with time lags), credit limit, sex, geographic region, credit score, among others. When I type margins, eyex(interest_rate) atmeans , I get the following results: ey/ex = -3.120104 Std.Err. = 6.181389 z-statistic = -0.50 P|>z|= 0.614 It surprises me that the p-value is 0.614 since I thought -margins- automatically excluded the covariates that are not significantly different from 0 (p>.10), but apparently it doesn't. Do you know if I should "clean up" my regression and exclude the non-significant covariates before I run -margins-? Many thanks, Sirenia * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/