Statalist The Stata Listserver


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

RE: st: Wald Chi-Square in Logistic with Cluster Option


From   "Clive Nicholas" <Clive.Nicholas@newcastle.ac.uk>
To   statalist@hsphsun2.harvard.edu
Subject   RE: st: Wald Chi-Square in Logistic with Cluster Option
Date   Tue, 14 Mar 2006 13:41:11 -0000 (GMT)

Daniel Indro replied:

> I also ran the regression without the cluster option.  The results gave
> me a reasonable chi-square.  However, some of the variables that I
> controled for did not come out significant.
>
> So, it seems to me that the problem could be due to clustering, or the
> inclusion of one particular year dummy variable, or both.  I checked
> Section 8.3 of Hosmer-Lemeshow book, have exhausted references to
> cluster option from the Stata website, and am wondering if someone could
> provide me with additional citations to help me learn more about this
> subject.  Also, what is the relationship between the number of
> observations within a cluster and the number of independent variables?
> Is there any specific requirement that the former be larger than the
> latter, or is this an irrelevant issue?

I'm glad you were able to find a solution to your previous solution,
despite my shamefully incorrect advice (thanks, Richard!).

I can't provide you with references right now, but I think I can answer
your question about clusters. The relationship between clusters and
independent variables is quite simple: you need to have more clusters than
there are IVs in your model. If you don't, your model should still run,
but Stata will report incomplete model diagnostics. Here's why, taken from
-whelp j_robustsingular-

**Are you using a svy estimator or did you specify the cluster() option?**

The VCE you have just estimated is not of sufficient rank to perform the
model test. As discussed in [R] test, the model test with clustered or
survey data is distributed as F(k,d-k+1) or chi2(k) where k is the number
of constraints and d=number of clusters or d=number of PSU's minus the
number of strata. Since the rank of the VCE is at most d and the model
test reserves one degree of freedom for the constant, at most d-1
constraints can be tested, so k must be less than d. The model that you
just fit does not meet this requirement.

Hope that helps.

CLIVE NICHOLAS        |t: 0(044)7903 397793
Politics              |e: clive.nicholas@ncl.ac.uk
Newcastle University  |http://www.ncl.ac.uk/geps

Whereever you go and whatever you do, just remember this. No matter how
many like you, admire you, love you or adore you, the number of people
turning up to your funeral will be largely determined by local weather
conditions.

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2020 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index