Dear list, I am running some zero inflated models and would like to know what the deviance of the models. Unlike running a normal GLM where the deviance is displayed in the summary all that is displayed in a summary of the zero inflated model is the log likelihood. I hope this isn't a read the manual question, and if it is I apologize for wasting your time, but if you could still send me a link of where I might find this information I would be very grateful! Thank you Lia [[alternative HTML version deleted]]
Lia McKinnon <l.mckinnon101 <at> gmail.com> writes:> I am running some zero inflated models and would like to know what the > deviance of the models. Unlike running a normal GLM where the deviance is > displayed in the summary all that is displayed in a summary of the zero > inflated model is the log likelihood. I hope this isn't a read the manual > question, and if it is I apologize for wasting your time, but if you could > still send me a link of where I might find this information I would be very > grateful! >are you looking for -2*logLik(model) ... ? This is the usual definition of the deviance (there are sometimes some subtle issues about the baseline model/additive constant). It would also be helpful to say what package you're using (pscl?), since zero-inflated models are not part of base R ... Ben Bolker
On Wed, 3 Apr 2013, Lia McKinnon wrote:> Dear list, > > I am running some zero inflated models and would like to know what the > deviance of the models. Unlike running a normal GLM where the deviance > is displayed in the summary all that is displayed in a summary of the > zero inflated model is the log likelihood.Zero-inflated models are not associated with a deviance in the GLM sense. They are a mixture of two models rather than a single model from an exponential family. Hence, zeroinfl/hurdle in "pscl" reports the log-likelihood but not the deviance. Possibly -2 times the likelihood is useful for you - as Ben already pointed out. Best, Z> I hope this isn't a read the manual question, and if it is I apologize > for wasting your time, but if you could still send me a link of where I > might find this information I would be very grateful! > > Thank you > Lia > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list > 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. >
The terminology could be specific to discipline. Check the Econometrics and Finance views. There isn't a health econometrics / statistics view yet which may have had this information, though you could look package "Design" and the author's web-page. Anirban Basu's and co-autors webpages at U Chicago may also contain some pointers and may be R code. Unfortunately, I have not used R for these type of models. HTH, Anupam. On 4/4/13, Lia McKinnon <l.mckinnon101 at gmail.com> wrote:> Dear list, > > I am running some zero inflated models and would like to know what the > deviance of the models. Unlike running a normal GLM where the deviance is > displayed in the summary all that is displayed in a summary of the zero > inflated model is the log likelihood. I hope this isn't a read the manual > question, and if it is I apologize for wasting your time, but if you could > still send me a link of where I might find this information I would be very > grateful! > > Thank you > Lia > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list > 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. >