Re: [R] glm(family=binomial(link=logit))

From: Peter Dalgaard <p.dalgaard_at_biostat.ku.dk>
Date: Sat 16 Jul 2005 - 01:58:51 EST

Robin Hankin <r.hankin@noc.soton.ac.uk> writes:

> Hi
>
> I am trying to make glm() work to analyze a toy logit system.
>
> I have a dataframe with x and y independent variables. I have
>
> L=1+x-y (ie coefficients 1,1,-1)
>
> then if I have a logit relation with L=log(p/(1-p)),
> p=1/(1+exp(L)).
>
> If I interpret "p" as the probability of success in a Bernouilli
> trial, and I can observe the result (0 for "no", 1 for "yes")
> how do I retrieve the coefficients c(1,1,-1)
> from the data?
>
> n <- 300
> des <- data.frame(x=(1:n)/n,y=sample(n)/n) # experimental design
> des <- cbind(des,L=1+des$x-des$y) # L=1+x-y
> des <- cbind(des,p=1/(1+exp(des$L))) # p=1/(1+e^L)
> des <- cbind(des,obs=rbinom(n,1,des$p)) # observation: prob of
> success = p.
>
>
> My attempt is:
>
> glm(obs~x+y,data=des,family=binomial(link="logit"))
>
> But it does not retrieve the correct coefficients of c(1,1,-1) ;
> I would expect a reasonably close answer with so much data.
>
> What is the correct glm() call to perform my logit analysis?

Apart from a sign error, the only fault seems to be that you think that 300 is a large number... Try upping n to 30000 and you'll see.

(The sign error is of course that log(p/(1-p)) = L implies that p = exp(L)/(1+exp(L)) = 1/(1+exp(-L))).

-- 
   O__  ---- Peter Dalgaard             ุster Farimagsgade 5, Entr.B
  c/ /'_ --- Dept. of Biostatistics     PO Box 2099, 1014 Cph. K
 (*) \(*) -- University of Copenhagen   Denmark          Ph:  (+45) 35327918
~~~~~~~~~~ - (p.dalgaard@biostat.ku.dk)                  FAX: (+45) 35327907

______________________________________________
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Received on Sat Jul 16 02:02:20 2005

This archive was generated by hypermail 2.1.8 : Fri 03 Mar 2006 - 03:33:45 EST