evaluate model performance when using 'withReplicates' in 'survey' package

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

evaluate model performance when using 'withReplicates' in 'survey' package

Hongwei Dong-2
Dear all,

I am using the 'withReplicates' function in the 'survey' package to estimate a zero-inflated Poisson model. The 'withReplicates' function gives me the coefficients, but no model performance indicators (e.g. log likelihood, pseudo R-squared, etc). Is there a way that I can evaluate the goodness-of-fit of a model when using 'withReplicates'?

My code looks like this:

zip_model<-withReplicates(design=mydesign, scale.weights=TRUE, quote(
                                                   coef(zeroinfl(Y ~ X1 + X2 + X3 | X1 + X2, weights=.weights))
                                                  ))
zip_model

I am aware that the 'fit.svyglm' function in the 'poliscidata' package can calculate R-squared for models that are estimated via 'svyglm'. Unfortunately, the 'svyglm' function estimates regular Poisson models only and cannot handle zero-inflated Poisson models.

Any advice is appreciated.

Best
Gary

        [[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.