[R] logistic regression by group?
noah at smartmediacorp.com
Thu Mar 4 20:22:12 CET 2010
Thanks for the quick reply.
I cant give any sample code as I don't know how to code this in R.
That's why I tried to pass along some pseudo code.
I'm looking for the best "beta" that maximize likelihood over all the
groups. So, while your suggestion is close, it isn't quite what I need.
I've seen the formula written as:
L = product( exp(xb) / sum(exp(xb)) )
Where sum(exp(xb)) represents the sum of all the items in the group.
Does that make sense?
On 3/4/10 4:04 AM, Corey Sparks wrote:
> Hi, first, you should always provide some repeatable code for us to have a
> look at, that shows what you have tried so far.
> That being said, you can use the subset= option in glm to subdivide your
> data and run separate models like that, e.g.
> fit.1<-glm(y~x1+x2, data=yourdat, family=binomial, subset=group==1)
> fit.2<-glm(y~x1+x2, data=yourdat, family=binomial, subset=group==2)
> where group is your grouping variable.
> Which should give you that kind of stratified model.
> Hope this helps,
> Corey Sparks, PhD
> Assistant Professor
> Department of Demography and Organization Studies
> University of Texas at San Antonio
> 501 West Durango Blvd
> Monterey Building 2.270C
> San Antonio, TX 78207
> corey.sparks 'at' utsa.edu
More information about the R-help