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: Significance of the variance of GLLAMM random intercept

From   Stas Kolenikov <>
Subject   Re: st: Significance of the variance of GLLAMM random intercept
Date   Mon, 5 Mar 2012 12:11:25 -0500

On Mon, Mar 5, 2012 at 11:12 AM, Maarten Buis <> wrote:
> On Mon, Mar 5, 2012 at 4:57 PM, junqing liu wrote:
>> I fitted a three level random intercept model using the gllamm. I
>> would like to test whether the variances of the random intercepts at
>> level 2 and level 3 are significant. But i am not sure how to conduct
>> the test. Any suggestions much appreciated.
> That is actually a very hard test as it is on "the boundary of the
> parameter space", that is, a variance can meaningfully take values
> larger than or equal to 0, so 0 is on the boundary of meaningful
> values for that parameter. Weird things happen there. See -help
> j_chibar- and
> Gutierrez, R., S. Carter, and D. M. Drukker. 2001. On boundary-value
> likelihood-ratio tests. Stata Technical Bulletin 60: 15–18.
> <>

I would also add that the standard technical reference is Self & Liang
(1987 JASA), see links at

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

*   For searches and help try:

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