Statalist


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

st: RE: Marginal Effect Probit - percent points?


From   "Villa Lora, Juan Miguel" <JUANMIGUELV@Contractual.iadb.org>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Marginal Effect Probit - percent points?
Date   Mon, 31 Aug 2009 10:55:36 -0400

I guess the interpretation would be as simple as "an unit change in the RHS variable would lead a change in the probability that the endogenous variable in X% ". Remember that this change is calculated on the probability average since you get a vector of dF/dX, which implies a % change of the cumulative density function.  

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Sophia Rüster
Sent: Lunes, 31 de Agosto de 2009 10:37 a.m.
To: statalist@hsphsun2.harvard.edu
Subject: st: Marginal Effect Probit - percent points?

Dear all,

I have run a Probit model and calculated the marginal effects afterwards.
Assuming a continuous RHS variable, the marginal effect is interpreted the following way: "An infinitesimal change in the RHS variable changes the probability that the endogenous variable takes the value of one by X%".

Are these X% (1) percent points or (2) percent of the actual value?

Example:
Pr(Y=1) = 0.5
Marginal Effect of x on y = 0.1 

(1) Does it mean Pr(Y=1) changes to 0.6 (0.5+0.1)
(2) or does it mean Pr(Y=1) changes to 0.55 (0.5*1.1)

Thanks a lot for a brief response!
Sophie

*
*   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/

*
*   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