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: xbu option in xtreg postestimation
From
Dorothy Bridges <[email protected]>
To
[email protected]
Subject
Re: st: xbu option in xtreg postestimation
Date
Sat, 7 May 2011 15:14:31 -0700
Good point! Thanks very much.
On Fri, May 6, 2011 at 5:25 AM, Scott Merryman <[email protected]> wrote:
> Since the intercepts do not change can't you get just carry the u
> values forward:
>
> webuse grunfeld,clear
> xtreg invest mval if year <1947, fe
> predict xb,xb
> predict u, u
> replace u = l.u if u == .
> gen xbu = xb+u
>
>
> Scott
>
> On Thu, May 5, 2011 at 9:27 AM, Dorothy Bridges <[email protected]> wrote:
>> Good morning,
>>
>> I am trying to fit the following three models to panel data:
>>
>> (1) y_it = a + Bt + e_it, by OLS
>> (2) y_it = a_i +Bt + e_it, with fixed effects
>> (3) y_it = a_i +Bt + e_it, with random effects
>>
>> The xtreg,fe and xtreg,re commands aren't working for me because (a)
>> as indicated above, I want to display the group-specific intercepts
>> rather than average them into one intercept, and (b) I need to make an
>> out-of-sample prediction that incorporates the group-specific means
>> (u_i, or a_i here), but the xbu option with "predict" does not allow
>> for out-of-sample predictions.
>>
>>
> *
> * 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/