[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: Running an empty model using xtmixed

From   "Yamamoto, Satomi" <>
To   "''" <>
Subject   RE: st: Running an empty model using xtmixed
Date   Fri, 28 Mar 2008 14:38:02 -0500


Thank you, Bobby.
I apprecaite your help.


-----Original Message-----
From: [] On Behalf Of Roberto G. Gutierrez, StataCorp
Sent: Friday, March 28, 2008 2:12 PM
Subject: Re: st: Running an empty model using xtmixed

Satomi Yamamoto <> asks:

> Could I use the xtmixed command to run a null model?  I used the
> pig.dta data ( and run

> xtmixed weight||id:, var

> But, I was not able to converge the data:


> Is there another way of running the intercept only model in Stata?
> Or, did I forget to include options in xtmixed? It would be great if
> somebody could help me wit this.

There is nothing that states you can't fit a null model with -xtmixed-, and the syntax you use is the correct one.  It is just that in your particular case -xtmixed- failed to converge, most likely because of a near-zero variance component at the -id- level.

    . sysuse auto, clear
    . xtmixed mpg || rep78:

does converge.  In fact, you can even be more "null" and do

    . xtmixed mpg

or if you want to get really crazy

    . xtmixed mpg, nocons

*   For searches and help try:

*   For searches and help try:

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