Statalist


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

Re: st: svy test


From   Richard Williams <Richard.A.Williams.5@ND.edu>
To   statalist@hsphsun2.harvard.edu, statalist@hsphsun2.harvard.edu
Subject   Re: st: svy test
Date   Wed, 10 Sep 2008 15:27:03 -0400

At 06:37 AM 9/10/2008, Maarten buis wrote:
--- Sahid Hocine <cahid2@googlemail.com> wrote:
> I am running a multivariate model from which  i would like to delete
> the non relevant variables.

You won't find much support for such stepwise procedures on this list
(with the possible exception of Rich Williams). The reasons for that
are summerized here:
http://www.stata.com/support/faqs/stat/stepwise.html

-- Maarten
I think stepwise can be useful as a diagnostic device, e.g. does mindless empiricism produce more or less the same model that your great theory does? If not, you might want to consider whether there is some other great theory you have overlooked. Also, I think there are situations where stepwise, while Satanic, is less Satanic than some of the alternatives. The present example doesn't seem to fall into either of these categories.

There are additional good reasons for not using sw with svy data. This FAQ explains them and suggests an alternative strategy:

http://www.stata.com/support/faqs/stat/stepsvy.html


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

*
* 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