[prev in list] [next in list] [prev in thread] [next in thread] 

List:       sas-l
Subject:    Re: WG: Confidence intervals calculated for multivariate logistic
From:       Dennis Gene Fisher <afdgf () UAA ! ALASKA ! EDU>
Date:       1999-06-29 21:52:11
[Download RAW message or body]

Jutta,
  There are four different options for obtaining different confidence
intervals in PROC LOGISTIC.  They all can go at the end of the
MODEL statement after a /   They are PLCL WALDCL PLRL
WALDRL  .  See which one gives you the confidence interval
that you want.

Dennis Fisher

Jutta.Friese@BC.BOEHRINGER-INGELHEIM.COM wrote:

>  Dear all,
>  I would like to ask your support on the following problem:
>
> > I ran a logistic regression for a response variable(two categories, e.g.
> > yes or no) in dependence from several parameters, e.g. age, sex, weight
> > and blood pressure. I wanted to plot the confidence intervals for the
> > response variable in dependence from one variable , e.g. blood pressure.
> > The values of the other variables are fixed, e.g. age=60 years, weight =
> > 60 kg, and sex = male.  Can I calculate these confidence intervals using
> > SAS?
> >
> Thanks in advance
>   Jutta

--
Dennis G. Fisher, Ph.D.
IVDU Project
University of Alaska Anchorage
3211 Providence Drive
Anchorage, AK 99508

[prev in list] [next in list] [prev in thread] [next in thread] 

Configure | About | News | Add a list | Sponsored by KoreLogic