Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

RE: st: Factor variables


From   "M.H.Hussein" <[email protected]>
To   "[email protected]" <[email protected]>
Subject   RE: st: Factor variables
Date   Sun, 17 Feb 2013 22:08:10 +0000

Thanks Richard.

In the command line, I have three variables and the interaction terms between each and a dummy (gt287). Using this command line I am getting the estimates for the coefficients for the variables when i.gt287=0 and the coefficients for the interaction terms.

To calculate the coefficients for the variables when i.gt287=1, I am adding the coefficients for each variable (when igt287=0) and the coefficient for the respective interaction terms (i.e. difference)

My question is how I can get the standard errors for the calculated coefficients-i.e when i.gt287=1?

Hope this is clearer.

Thanks again,
Mohamud
________________________________________
From: [email protected] [[email protected]] on behalf of Richard Goldstein [[email protected]]
Sent: 17 February 2013 21:47
To: [email protected]
Subject: Re: st: Factor variables

I do not find this entirely clear; however, if you look at the help file
(-h fvvarlist-), under the "base levels" section, you will see that
"ibn." means "no base level" and this might be what you want

do realize, however, that you are changing the null hypotheses for your
tests

Rich

On 2/17/13 4:39 PM, M.H.Hussein wrote:
> Hello Statalisters
>
> I am running the regression command line below which contains factor variables with interactions. The command line returns coefficients for all variables in the model when the indicator gt287=0 and interaction terms.
>
>  xtreg  TCOST  i.gt287##c.P_O i.gt287##c.Y_TCOST10 i.gt287##c.enforcement10, fe
>
>
> I would to obtain the estimates for coefficients when gt287=1 alongside above estimates, so that I can get standard errors and t statistics for all coefficients.
>
> Could you tell me if this possible and if so how I can write the command to get all estimates in a single command line?
>
> Thanks,
> Mohamud
> -- Kentforlife.net - the email service for alumni of the University of Kent
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/

-- Kentforlife.net - the email service for alumni of the University of Kent


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index