Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: probit mean


From   Marcello Pagano <pagano@hsph.harvard.edu>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: probit mean
Date   Mon, 13 Nov 2006 12:46:55 -0500

I don't know about what was meant by the first part, but the second part "My colleague estimates a probit model and then puts mean projections into the explanatory variables..." is a silly way to proceed. Think about it. You would be finding the probability for the "average" person, i.e. one with a gender of 0.46, for example. It is far better to average out the probabilities.

m.p.

Dirk Nachbar wrote:

All

A colleague received a reply from an academic saying " Remember that the probit does not predict at the mean and an adjustment needs to be made for this." Unfortunately I couldn't tell him what she meant by that. My colleague estimates a probit model and then puts mean projections into the explanatory variables to predict a dependent probability.

Does any of you know what the academic meant by that?

Dirk Nachbar
Assistant Economist
WWEG FPWD CPSA
Department for Work and Pensions
Level 2, The Adelphi
1-11 John Adam St
WC2N 6HT London
020 796 28531 uk

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index