Re: st: Hurdle models vs. zero inflated models

[Austin Nichols <austinnichols@gmail.com>]

Pooja Desai <desai.pooja87@gmail.com>:

There is no guarantee that adding structure with a hurdle model or a
zero inflation component will improve your predictions, and you might
consider starting with -poisson- or -glm- with log link, which require
fewer assumptions for consistent estimation:

http://blog.stata.com/2011/08/22/use-poisson-rather-than-regress-tell-a-friend/
http://www.stata.com/meeting/boston10/boston10_nichols.pdf

Note in particular that a large number of zeros is *not* sufficient to
justify using a hurdle model or a zero-inflated model, since
predictors may generate exactly this distribution of outcomes, with a
preponderance of very low predicted values (that round to zero).  A
better reason to move to a 2-part model is a theoretical argument that
different predictors matter in different processes.

On Mon, Nov 18, 2013 at 4:20 AM, Pooja Desai <desai.pooja87@gmail.com> wrote:
> Hello All,
>
> I have a variable which represents the length of stay in the hospital
> for patients. 90% of the patients were not hospitalized so they have a
> 0 length of stay. I was considering using either a zero-inflated
> negative binomial regression or a hurdle model (logit and zero
> truncated negative binomial) for this variable. Since all the zeros on
> the length of stay come from one source—the patients not having a
> hospitalization, I thought a hurdle model would be a better option
> (there are no sampling zeros). Is that an appropriate choice?
>
> If so, I need to find the predicted length of stay from the model. Is
> there a way to get the predicted value after accounting for both the
> logistic and zero truncated negative binomial models?

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/