Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: RE: RE: out-of-sample predictions with GLM


From   jc1926@gmx.de
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: RE: out-of-sample predictions with GLM
Date   Thu, 26 Jul 2007 16:09:19 +0200

Hello,

Many thanks for the follow-up. I do have a situation where the proportions should add up to one, as in the city budget example.

In my case, my unit of observation is the share of a new car model as a percent of the total market for new cars in a given year. I have this share over several years. Because there are hundreds of car models in my data, I don't think it would be feasible to use a multinomial logit. None of my observation are zero or one. The x's are mostly car attributes for the particular car models, many of which change over time.

I pool all the data and estimate with:
glm share_i_year_t x_i_t, fam(binomial) link(logit) robust

Company dummies are included as fixed effects. The predicted shares from the model sum to 1 for a given year, which is gratifying. Out-of-sample predictions do not sum to one, which is not gratifying but also not unexpected. Maarten's comment makes me wonder if there is a better approach.

Any further insights would be welcome.



-------- Original-Nachricht --------
Datum: Thu, 26 Jul 2007 15:17:43 +0200
Von: "Maarten Buis" <M.Buis@fsw.vu.nl>
An: statalist@hsphsun2.harvard.edu
Betreff: st: RE: RE: out-of-sample predictions with GLM

> ---I wrote:
> > A situation where you want proportions to add up to 1
> > within an observation occurs when you have for each
> > observation multiple proportions, e.g. for each city
> > you have the proportions of the city budget spent on
> > safety, education, transport, etc. In this case the
> > proportions within a city should add up to 1. A model
> > that deals with this situation is -dirifit-,see:
> > http://home.fsw.vu.nl/m.buis/software/dirifit.html 
> 
> A couple of months ago I posted a fractional multinomial 
> logit -fmlogit- on the statalist 
> http://www.stata.com/statalist/archive/2007-05/msg00449.html
> 
> -fmlogit- relates to -dirifit- like fractional logit relates 
> to beta regression. 
> 
> Can you tell us a bit more about your problem? What are
> the dependent and independent variables? Do you have many 
> observations with a proportion equal to 0 or 1. Papke and 
> Wooldridge are overly optimistic about the ability of
> dealing with those problems using the fractional 
> (binomial or multinomial) logit model.
> 
> Hope this helps,
> Maarten
> 
> -----------------------------------------
> Maarten L. Buis
> Department of Social Research Methodology 
> Vrije Universiteit Amsterdam 
> Boelelaan 1081 
> 1081 HV Amsterdam 
> The Netherlands
> 
> visiting address:
> Buitenveldertselaan 3 (Metropolitan), room Z434 
> 
> +31 20 5986715
> 
> http://home.fsw.vu.nl/m.buis/
> -----------------------------------------
> 
> 
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/

-- 
Psssst! Schon vom neuen GMX MultiMessenger gehört?
Der kanns mit allen: http://www.gmx.net/de/go/multimessenger
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index