[Basta-users] Varying only some parameter estimates across categorical covariates

[Fernando Colchero]

Dear Andrea,

   Unfortunately BaSTA cannot do that yet. However, you can use the Kullback-Leibler discrepancies to determine which parameters may differ between categorical covariates. I agree it’s not the most parsimonious way of doing it. We want in the future to include that option. We’ll let you know.

Best,

Fernando
_________________________________

Fernando Colchero
Associate Professor
Department of Mathematics and Computer Science
Biodemography Unit

Tlf.              +45 65 50 23 24
Email          colchero at imada.sdu.dk <mailto:colchero at imada.sdu.dk>
Web            www.sdu.dk/staff/colchero <http://www.sdu.dk/staff/colchero>
Pers. web   www.colchero.com <http://www.sdu.dk/staff/colchero>
Adr.             Campusvej 55, 5230, Odense, Dk

University of Southern Denmark
_________________________________




> On 02 Apr 2018, at 16:32, Andrea Grunst <asgrunst at gmail.com> wrote:
> 
> Hello,
> 
> I have a question regarding whether it is possible to vary only some parameter estimates across levels of a categorical covariate when running BaSTA.  For instance, when working with a Weibull bathtub model, is it possible to vary only b0 and b1 across categories, but not a0, a1 and c?  I am asking because only b1 (and maybe b0) appear to differ between the groups that I am considering, so it seems superfluous to also estimate the other parameters separately (this will elevate the DIC).  Any suggestions would be appreciated.
> 
> Thanks you in advance for your help.
> 
> Andrea Grunst
> _______________________________________________
> Basta-users mailing list
> Basta-users at lists.r-forge.r-project.org
> https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/basta-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.r-forge.r-project.org/pipermail/basta-users/attachments/20180403/c2db2774/attachment.html>