Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: prediction


From   Richard Williams <Richard.A.Williams.5@ND.edu>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: prediction
Date   Fri, 05 Aug 2005 11:13:39 -0500

At 11:56 AM 8/5/2005 -0400, Raymond wrote:
if I run a regression without the dummies, then the coefficients of
the rest variables will be different.
What I need to do is using the coefficient of the independent
variables calculated while gender is included to estimated the
dependent variable while gender is exluded from the prediction
equation.
First off, you probably don't want to do that. You would be using parameter estimates that no model has actually generated. I very strongly suspect that what you propose doing is the wrong way to go but then you know your work better than I do. (Students occasionally want to compute predicted values based on some hypothetical set of values, and they ask me whether the non-significant coefficients should just be treated as zero; I say no, either leave them as is or re-estimate the model with those variables dropped.)

But, if you are bound and determined to do it anyway, see this recent message from Vince Wiggins:

http://www.stata.com/statalist/archive/2005-07/msg00617.html

You could just modify the foo.ado program he shows to set the coefficients to 0.

Yet another possibility would be to do a -gen- command yourself, leaving out the dummy vars; or do a -predict- command followed by a -replace- where you subtract the contribution created by the dummies.


-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX: (574)288-4373
HOME: (574)289-5227
EMAIL: Richard.A.Williams.5@ND.Edu
WWW (personal): http://www.nd.edu/~rwilliam
WWW (department): http://www.nd.edu/~soc

*
* 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