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, is already up and running.

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

Re: st: suest equivalent for areg.

From   Austin Nichols <>
Subject   Re: st: suest equivalent for areg.
Date   Fri, 20 May 2011 17:13:24 -0400

Then confirm you get the same answer with a stacking approach:
noting that clustering by id will subsume clustering by obs.

On Fri, May 20, 2011 at 4:32 PM, Schaffer, Mark E <> wrote:
> Sumedha,
> Why not partial out the fixed effects by hand and then use -regress-?  An easy way to do this is with Ben Jann's -center- command.  Then do your estimations on the transformed data.
> If you're planning on using -suest-, you'll probably need to use the undocumented -dof- option of -regress- when you do your estimations before invoking -suest-.  This is because you will have to tell -regress- that you've lost X degrees of freedom because of the X individual effects.  The syntax is easy, e.g.,
> regress y x1 x2, dof(25)
> tells -regress- that there are 25 degrees of freedom, i.e., the macro e(df_r) will have 25 in it.
> HTH,
> Mark
>> -----Original Message-----
>> From:
>> [] On Behalf Of
>> Gupta, Sumedha
>> Sent: 20 May 2011 17:34
>> To:
>> Subject: RE: st: suest equivalent for areg.
>> Hi Jorge,
>> Thank you for your reply but I am running into exactly the
>> problem you anticipated. Suest doesn't allow areg and xtreg
>> and running ols with individual dummies is not an option with
>> svy  as that does not allow vce (cluster id).
>> Does anybody else have any thoughts on this please?
>> Many thanks.
>> Sumedha.

*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index