Notice: On March 31, it was **announced** that Statalist is moving from an email list to a **forum**. The old list will shut down on April 23, and its replacement, **statalist.org** is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

From |
Rajan Panta <rajan.panta@anu.edu.au> |

To |
"statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |

Subject |
st: Selectivity bias in Multinomial logit and selmlog command |

Date |
Fri, 31 May 2013 06:08:12 +0000 |

Dear Statalist users, Version: Stata /SE 11.2 for windows I am using the user written -selmlog- command (version 13, August 2006) to correct for the selection bias in the multinomial logit selection equation with 3 categories. ( I downloaded the .ado file from http://www.parisschoolofeconomics.com/gurgand-marc/selmlog/selmlog13.html which is based on the paper "Selection Bias Corrections Based on the Multinomial Logit Model: Monte-Carlo Comparisons", François Bourguignon, Martin Fournier, Marc Gurgand, Journal of Economic Surveys, vol. 21, no.1, 2007). The explanatory variables in the selection equation are all categorical variables and there are two more nominal explanatory variables in the main equation. I ran the following code. selmlog depvar varlist1, select(Multinomial logit categories=varlist2) dmf(2) wls boot(50) mloptions(rrr level (95)) gen(rho1) I ran the regression for each categories creating missing values for dependent variables in the main equation other than that particular category. I have following questions: 1. I get the series of error messages such as -Beware: for 885 observations, the computed heteroskedastic residual variance is negative-. I get the output after the error messages, but I do not understand the warning messages and how to address it. 2. What is the criteria to choose the number in the dmf option (0, 1 or 2) or how many selection correction terms should be there in the output if I have 3 categories in the multionomial logit selection equation? In some papers I see only two different selection terms for the output depending on the categories. 3. How do we interpret the coefficients of the variables and of the selection terms (sigma 2, rho0, rho1 and rho2) in the “Anciliary” section of output. Any suggestions will be appreciated. Best regards, Rajan * * 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/

- Prev by Date:
**st: NonLinear GMM using moment evaluator program** - Next by Date:
**st: Stata Summer School in Stockholm** - Previous by thread:
**st: NonLinear GMM using moment evaluator program** - Next by thread:
**st: Stata Summer School in Stockholm** - Index(es):