R help - Aug 2009 - Logit Model... GLM or GEE or ??

Posted about this earlier.  Didn't receive any response

But, some further research leads me to believe that MAYBE a GLMM or a 
GEE function will do what I need.

Hello,

I have a bit of a tricky puzzle with trying to implement a logit model 
as described in a paper.

The particular paper is on horseracing and they explain a model that is 
a logit trained "per race", yet somehow the coefficients are combined 
across all the training races to come up with a final set of coefficients.

My understanding is that they maximize log likelihood across the entire 
set of training races. Yet this isn't just as standard logit model as 
they are looking at data "per race".

This is a bit hard to explain, so I've attached a tiny pdf of the 
paragraph from the paper explaining this.

Like everything else in the data/stat/econ world, there is probably a 
library in R that does this kind of thing, but after 3 days of heavy 
google research, I've been unable to find it.

Does anyone have any suggestions??

Thanks.

-N


Attached is a jpg of the book page describing what I'm trying to do...

You could take a look at:

M West, PJ Harrison, HS Migon - Journal of the American Statistical  
Association, 1985 - jstor.org
Page 1. Dynamic Generalized Linear Models and Bayesian Forecasting

and the subsequent literature it has generated... or along the same  
lines the literature on chess
ratings.


url:    www.econ.uiuc.edu/~roger            Roger Koenker
email    rkoenker at uiuc.edu            Department of Economics
vox:     217-333-4558                University of Illinois
fax:       217-244-6678                Urbana, IL 61801



On Aug 6, 2009, at 5:00 PM, Noah Silverman wrote:
> Posted about this earlier.  Didn't receive any response
>
> But, some further research leads me to believe that MAYBE a GLMM or  
> a GEE function will do what I need.
>
> Hello,
>
> I have a bit of a tricky puzzle with trying to implement a logit  
> model as described in a paper.
>
> The particular paper is on horseracing and they explain a model that  
> is a logit trained "per race", yet somehow the coefficients are  
> combined across all the training races to come up with a final set  
> of coefficients.
>
> My understanding is that they maximize log likelihood across the  
> entire set of training races. Yet this isn't just as standard logit  
> model as they are looking at data "per race".
>
> This is a bit hard to explain, so I've attached a tiny pdf of the  
> paragraph from the paper explaining this.
>
> Like everything else in the data/stat/econ world, there is probably  
> a library in R that does this kind of thing, but after 3 days of  
> heavy google research, I've been unable to find it.
>
> Does anyone have any suggestions??
>
> Thanks.
>
> -N
>
>
> Attached is a jpg of the book page describing what I'm trying to do...
>
>
> ______________________________________________
> 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.