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: three-level xtreg

From   Stas Kolenikov <>
Subject   Re: st: three-level xtreg
Date   Wed, 29 Jun 2011 12:10:21 -0400

A random effect model would be estimable with

xtmixed incorr spanish otherl bmi dep bmi_p50 bmi_p80 popbmi || aid: || scid:

For a fixed effects model, you need to decide which level you will be
having your fixed effects at, as it does not make sense to have more
than one level. If you have them at personal level, then the school
effects are not identified (think about the dummy variable
interpretation: the school dummy is the sum of the student dummies).

On Wed, Jun 29, 2011 at 12:05 PM, Lisa Marie Yarnell
<> wrote:
> Hello Statalisters.  My question is on three-level xtreg models.
> I have seen 2-level xtreg models (e.g., time points nested within persons):
> xtreg incorr spanish otherl bmi dep bmi_p50 bmi_p80 popbmi, i(aid) fe
> I want to nest time points within persons within schools.  I got an error message when I tried:
> . xtreg incorr spanish otherl bmi dep bmi_p50 bmi_p80 popbmi, i(aid scid) fe
> i():  too many variables specified
> What is the code for this?
> Thanks for your help,
> Lisa
> *
> *   For searches and help try:
> *
> *
> *

Stas Kolenikov, also found at
Small print: I use this email account for mailing lists only.

*   For searches and help try:

© Copyright 1996–2016 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index