Statalist


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

st: re: predicting after probit (was: MIME-Version: 1.0)


From   Kit Baum <baum@bc.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: re: predicting after probit (was: MIME-Version: 1.0)
Date   Thu, 27 Nov 2008 07:47:38 -0500

< >	
Gabriele wrote

I do a probit regression in a cross section of individuals that are
from many countries (I have abouth one thousand of observations for
each country). For each individual I have a variable where it's
indicated his country. I desire to obtain a predicted  probability
for the dependent variable   not at individual's level but at county
level. How can i do?


In the enclosed, imagine that the individual observations for each company belong to individuals. Your 'country' is my 'company', the group with which the 'individual' is affiliated.

webuse grunfeld,clear
bys company: egen invmu = mean(invest)
bys company: g hi = invest > invmu
probit hi mvalue kstock
predict double hihat, pr
bys company: egen prcomp = mean(hihat)
list invest hi hihat prcomp if company < 3


Kit Baum, Boston College Economics and DIW Berlin
http://ideas.repec.org/e/pba1.html
An Introduction to Modern Econometrics Using Stata:
http://www.stata-press.com/books/imeus.html


*
*   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/



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