Statalist


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

st: clogit with cluster vs. not


From   "C.Chetkiewicz" <c.chetkiewicz@comcast.net>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: clogit with cluster vs. not
Date   Fri, 2 May 2008 12:32:35 -0700

Hello,

I am using Stata 9.2.

I am running clogit on a 1:20 matched study for telemetry steps (sampled_path) as mentioned previously.  I am clustering on individual animal to produce robust SEs.  

I noticed that when I use the cluster option, I get the Wald Chi output for my model and when I don't cluster I get a LR chi2 output. Can someone briefly explain why these change with cluster options or point me to a discussion/overview of this?  

Are there implications in reporting one over the other?  I noticed I get the LR chi2 from -fitstat- after -clogit-.  Biologically, clustering on individual animal makes much more sense so I would prefer to use this.  

Either way, these example models are significant correct?  

Best,

Cheryl


*******************************************
OUTPUT

1. clogit   sampled_path rdens2k_max trees_mean nonveg_mean, group( uidstep) cluster( uidpath) nolog

Conditional (fixed-effects) logistic regression   Number of obs   =      11613
                                                  Wald chi2(3)    =      72.70
                                                  Prob > chi2     =     0.0000
Log pseudolikelihood = -1673.8298                 Pseudo R2       =     0.0058

                                (Std. Err. adjusted for 5 clusters in uidpath)
------------------------------------------------------------------------------
             |               Robust
sampled_path |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
 rdens2k_max |   .2901546    .105544     2.75   0.006     .0832921     .497017
  trees_mean |   .4624635   .3109852     1.49   0.137    -.1470563    1.071983
 nonveg_mean |   .9611335   .8892528     1.08   0.280      -.78177    2.704037
------------------------------------------------------------------------------

2.  clogit   sampled_path rdens2k_max trees_mean nonveg_mean, group( uidstep) nolog

Conditional (fixed-effects) logistic regression   Number of obs   =      11613
                                                  LR chi2(3)      =      19.58
                                                  Prob > chi2     =     0.0002
Log likelihood = -1673.8298                       Pseudo R2       =     0.0058

------------------------------------------------------------------------------
sampled_path |      Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
-------------+----------------------------------------------------------------
 rdens2k_max |   .2901546   .0774408     3.75   0.000     .1383734    .4419357
  trees_mean |   .4624635   .2151835     2.15   0.032     .0407116    .8842154
 nonveg_mean |   .9611335   .3174182     3.03   0.002     .3390053    1.583262
------------------------------------------------------------------------------



***********************************************
Cheryl Chetkiewicz
P.O. Box 60208
Palo Alto, CA, 94306
USA
Telephone: 650-331-0069
E-Mail: c.chetkiewicz@comcast.net
 please consider the environment before printing this email



*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index