Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

st: marginal effects after cmp mprobit

From   Sarit Weisburd <>
Subject   st: marginal effects after cmp mprobit
Date   Wed, 27 Jun 2012 12:59:44 +0300

Hello statalist,

I have been using cmp to estimate a multinomial probit with an
instrument. While the output includes coefficients for _outcome_1_2
and _outcome_1_3. I cannot figure out how to compute marginal effects.
I have tried mfx, mfx2 and margins - but they all result in errors

This is the cmp command:
cmp (o_accident = c_insurance car_value01 engine year /*
  */  owner_sex distance bagrut avg_inc year_0* d_N* d_S* /*
  */   policy_length mean_* daccident1) (c_insurance=company), /*
  */ ind($cmp_mprobit $cmp_cont) nolr cluster(client_num) tech(dfp nr)

These are Stat's responses to my attempt at postestimation:
mfx2 - "there is only one outcome in o_accidents"
margins / mfx, predict(pr outcome(1)) nonlinear - "Equation #1 is not
ordered probit. outcome() is not allowed".

I would be happy to hear any suggestions,

Sarit Weisburd

Doctoral Candidate
Department of Economics
Hebrew University
Mount Scopus Campus
Jerusalem 91905, Israel
*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index