Statalist The Stata Listserver

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

Re: stepwise like SPSS? [was: RE: st: Means test]

From   Richard Williams <>
Subject   Re: stepwise like SPSS? [was: RE: st: Means test]
Date   Mon, 19 Feb 2007 06:59:03 -0500

At 02:22 PM 2/16/2007, David Zatz wrote:
Doing this, I end up with exactly the same results SPSS provides. This is
wondrous - as is the bestreg program that I will be experimenting with next
- as it will allow me to drop SPSS (costly and clunky, and can't be used on
my current Mac) for Stata.
One thing that SPSS does that Stata does not: SPSS gives you information on Variables not in the equation, e.g. it will tell you the coefficient and T-value each variable would have if added to the model next. This can be useful for determining whether the variables that made it in were clear winners, or narrow ones. If the latter, then that increases the possibility that a slightly different sample could have produced a different set of winners.

In the absence of such a feature, I might be tempted to run a 2nd stepwise regression, where I omitted all the variables that made it in the first time. If the "losers" produced a model that fit almost as well, I'd have less confidence in my stepwise results. Not that I'd be wildly confident in them otherwise, but it would be nice to know if a purely empirical approach for selecting variables produces clear winners or narrow ones.

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):
WWW (department):
* For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index