Statalist


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

st: can I get standardized coefficients by standardize variables first?


From   Mandy fu <mandy.fu1@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: can I get standardized coefficients by standardize variables first?
Date   Tue, 3 Mar 2009 12:50:39 -0500

Hi all,

I'm running some panal data. So the commands I 'll use are something
like -xtreg , fe-, or -xtreg,re-. Since the beta coeffients
(standardized coefficients) are not an option for xtreg command, I
tried to standardize all the variables in the whole data set first
and then use the panel data methods on the standardized variables:
-------------
.local var y x1 x2 x3 x4
.egen new_`var'=std(`var')
.xtreg `var',fe
---------------
But the results are different from using the beta coefficients got
from an OLS regression adding dummy variables to indicate  the group
variable i for the panel data.
-----------------------------------------
**dummy1,dummy2,..dummyn are the dummy vriables based on group i.
.reg y x1 x2 x3 x4 dummy*,b
-----------------------------------------
Could someone give me some hints why the results from these two ways
are different? Is it a correct way to get beta coefficents by
standardized variables first and then use the regular commands?

Thanks a lot in advance!

Mandy
*
*   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