Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: SVYreg Command

From   Rufus Browning <[email protected]>
To   [email protected]
Subject   Re: st: SVYreg Command
Date   Wed, 15 Oct 2003 18:06:24 -0700

At 02:46 PM 10/15/2003 -0400, Alison wrote:

I am using the svy commands for the first time, and am having some troubles.
First of all -- I need to have the svyreg regressions produce beta
coefficients. The output does not specify if in fact the coefficients are
standardized or not, and the manuals are not straightforward about this issue.
The beta option for the regress command does not work for the svyreg command.
So ... are the coefficients standardized, and if not, how would I calculate
This might meet your needs. You can enter a p-weight variable into an ordinary regression command in brackets and specify the beta option:

. regress y x [pw = yourwtvar] , beta

-regress- will take your p-weights into account and still produce the betas (standardized coefficients).

-regress- also permits clusters and robust variance estimates, but the help file says the cluster option may not be combined with beta. See -help regress-.

Rufus Browning
Public Research Institute, San Francisco State University

I tried using svymean to generate the standard deviations, and use the old
formula (beta = (sx/sy)*b) to calculate the betas, but svymean produces the
standard error for the mean, and not the standard deviation -- this is at
least how the output is labeled, and the manuals once again are not
straightforward in telling me what is being reported.  So ... what would I do

Finally, I am using version 7 of STATA -- I do not presently have access to
STATA8, but I could possibly get access.  Does STATA8 have a svyreg command
option that produces the standardized regression coefficients?  If so, I will
try to get STATA8.

Thank you in advance for all of your help,


*   For searches and help try:
*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index