s-news
[Top] [All Lists]

Re: logistic regression

To: <s-news@lists.biostat.wustl.edu>
Subject: Re: logistic regression
From: "Umberto" <anglagard77@libero.it>
Date: Thu, 31 Jul 2003 19:37:50 +0200
References: <3F29397D.6030906@amc.uva.nl>


> Hi,
>
> I'm new to logistic regression and therefore this stupid question:
>
> What I have understand from logistic regression is that the binary
> response variable is transformed to log(p/(1-p)).
> 1) How is the fraction p calculated for an individual response value (0
> or 1). Probably, if for a certain predictor value only 1 response is
> available the fraction is aslo always 1 or 0, or is the fraction p
> calculated a pre-specified range for a predictor value?
>
if you have a set of p covariates end their effects on the variable response
Y={0;1} are sintetized by the regression coefficients beta (where beta is a
p dimensional vector) you can write
logit(p/(1-p))=X*beta
where p=Prob(Y=1)

All the major statistical packages (Sas, S-plus ecc) can compute the beta
estimates (through iterative alghoritms), say b*, so the estimates for p,
say p*, are given by
p*=exp(Xb*)/(1+exp(Xb*))
or equivalenty
log(p*/(1-p*))=Xb*



<Prev in Thread] Current Thread [Next in Thread>