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 at the end of May, and its replacement, statalist.org is already up and running.


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

Re: st: Standardized Coefficients after Regress with Cluster Adjustment


From   "Roberto Liebscher" <roberto.liebscher@ku-eichstaett.de>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Standardized Coefficients after Regress with Cluster Adjustment
Date   Mon, 18 Jun 2012 14:22:44 +0200 (CEST)

I think I got the point. The problem is that we use an estimate for the
standard deviation of each variable and assume that this standard
deviation stems from the same population. But this only makes sense if the
standard deviation in each cluster is the same.

Have I got you right?

Thanks for your kind help.

Roberto

> On Mon, Jun 18, 2012 at 9:28 AM, Roberto Liebscher wrote:
>> When you want to get the standardized coefficients you are usually
>> allowed
>> to add the option -beta- to the regress command in stata. However this
>> does not work when you use the option -cluster(group)- to adjust the VCE
>> for cluster correlated disturbances.
>>
>> What is the reason for this and is it allowed to calculate the
>> standardized coefficients by hand in this case? Unfortunately I can not
>> understand why this should not be allowed as standardization is just a
>> transformation of the coefficients.
>
> You are allowed to do whatever you like. However, with such freedom
> comes responsibility. It is now up to you to convince others that what
> you did actually makes sense.
>
> To me the transformation implicit in standardized coefficients makes
> (some) sense when there is a meaningful estimate of the standard
> deviations for each variable. If that is in doubt, e.g. because you
> have a clustered sample, than standardized coefficients cease to be
> meaningful.
>
> Hope this helps,
> Maarten
>
> --------------------------
> Maarten L. Buis
> Institut fuer Soziologie
> Universitaet Tuebingen
> Wilhelmstrasse 36
> 72074 Tuebingen
> Germany
>
>
> http://www.maartenbuis.nl
> --------------------------
> *
> *   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