Statalist


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

Re: st: Mean-Centering predictor variables in Probit regression?


From   "Austin Nichols" <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Mean-Centering predictor variables in Probit regression?
Date   Fri, 7 Dec 2007 13:11:08 -0500

Kam Kup <kamkupan@yahoo.com>:
1. It is sometimes harder to interpret coefs.
2. If you are estimating the mean and sd for each var, then you are
using generated regressors and should adjust the VCE.

But, short answer: no, there's no reason I know of not to do it.

On 12/7/07, Kam Kup <kamkupan@yahoo.com> wrote:
> Is it accepted to standardize covariates in a discrete
> choice model in order to avoid multicollinearity.
> (Standardizing here refers to mean-centering each
> variable and then divide it by its standard
> deviation).
>
> I know that this will yield different estimates in the
> non-linear case.  But my question is, are there any
> good reasons not to do it?
>
> Thank you!
>
> Kam
*
*   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 1996–2021 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index