Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

RE: st: RE: Re: standardized betas in Prais-Winsten Regression


From   Schöler, Lisa <lschoeler@wiwi.uni-frankfurt.de>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   RE: st: RE: Re: standardized betas in Prais-Winsten Regression
Date   Tue, 26 Oct 2010 18:53:10 +0000

Hi Statalist,

I am sorry that I have a question again, but I can't figure out my problem.

As suggested by Kit, I z-transformed my independent and dependent variables. I used the command

.center varlist, standardize

Now my R^2 and F value go down. How can that happen, I have learned that R^2 and F value don't change if you z-transform.

If I ran a regular OLS instead of a Prais-Winsten regression, R^2 and F value don't change. So I guess, it has something to do with the Prais-Winsten regression. Does anybody know the reason?

Thank you very much
Lisa

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Christopher Baum
Sent: Saturday, October 23, 2010 1:42 AM
To: statalist@hsphsun2.harvard.edu
Subject: re: st: RE: Re: standardized betas in Prais-Winsten Regression

<.>
Lisa said

I centered all my variables and ran the Prais-Winsten Regression with the centered variables. The coefficients are the same for uncentered and centered variables. I think that this can't be true. Does anyone know why this problem occurs?


Actually if you want beta coefficients, you want to not only center them but z-transform them, that is, subtract the mean and divide by the sd. Subtracting constants from y and x doesn't change their correlation.

But z-transforming is a linear transformation, and regression of a lin.transf.(y) on a lin.transf.(x) doesn't change r^2. F, t-stats, etc. So why should this be surprising?

Kit

Kit Baum   |   Boston College Economics & DIW Berlin   |   http://ideas.repec.org/e/pba1.html
                              An Introduction to Stata Programming  |   http://www.stata-press.com/books/isp.html
   An Introduction to Modern Econometrics Using Stata  |   http://www.stata-press.com/books/imeus.html


*
*   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   |   Site index