Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Francesca Fabbri" <Francesca.Fabbri@lrz.uni-muenchen.de> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: mtreatreg - marginal effects |
Date | Sat, 12 Feb 2011 18:10:45 +0100 (CET) |
Hi, I am using mtreatreg (by Partha Deb) to estimate multinomial treatment effects on a continuous and on a bivariate outcome. I have 2 questions regarding the procedure. - In the original paper where Deb and Trivedi develop the multinomial treatment for a negative binomial ("Specification and simulated likelihood estimation of a non-normal treatment-outcome model with selection: Application to health care utilization", Econometrics Journal (2006), volume 9, pp. 307?331), the authors explain that they calculate the marginal effects using Montecarlo. Is this the case for mtreatreg as well, when the outcome variable is a probit? - And is this the reason why Stata would not calculated the standard errors of the marginal effects? When I run mfx after mtreatreg to obtain marginal effects from the multinomial+probit specification, Stata would not calculate the standard errors of the marginal effects, giving the error: "warning: default predict() is unsuitable for standard-error calculation; option nose imposed" Thanks in advance, Francesca Francesca * * 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/