[R-sig-ME] Binary clustered data with negative intraclass correlation

[Jarrod Hadfield]

Hi,

Do you have two observations per cluster, as in the paper you cite?

Cheers,

Jarrod


Quoting Henrik Hermansson <ajhhermansson at gmail.com> on Fri, 6 Apr 2012  
15:16:19 +0100:

> Dear all,
>
> I have been searching far and wide for a solution to the following problem:
>
> I have clustered data with a binary outcome. But, rather unusually
> (but logically, given the application) the correlation within the
> clusters is negative, as a result of competition within the cluster. I
> understand from Ten Have et al. (1998) that in such cases, most
> clustered-standard-errors estimators will be biased, since they assume
> a positive intraclass (intracluster) correlation.
>
> Thomas R. Ten Have, Allen Kunselman & Euginia Zharichenko (1998):
> Accommodating negative intracluster correlation with a mixed effects
> logistic model for bivariate binary data, Journal of Biopharmaceutical
> Statistics, 8:1, 131-149
>
> I'm hoping that someone will be able to point me towards a R package
> (or similar) which can perform random effect clustered standard errors
> regression without the limitation of an assumed positive intraclass
> correlation.
>
> Many thanks in advance for any help and guidance!
>
> Best regards
> Henrik Hermansson
> PhD Candidate
> Department of Political Science
> Trinity College Dublin
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>
>



-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.