Date:         Tue, 13 Feb 2007 02:22:03 -0800
Reply-To:     Albert-jan Roskam <fomcl@YAHOO.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Albert-jan Roskam <fomcl@YAHOO.COM>
Subject:      Re: PROC LOGISTIC confidence limits
Comments: To: "N. Knese" <n.knese@GMX.DE>
In-Reply-To:  <20070213100441.132540@gmx.net>
Content-Type: text/plain; charset=iso-8859-1

Hi,

You may use something like this, where 'estimates' are
your betas:

OR    = exp(estimate);
lo_OR = exp(estimate - 1.96*stderr);
up_OR = exp(estimate + 1.96*stderr);

Note that this calculates 95% CIs.

Cheers!
Albert-Jan



--- "N. Knese" <n.knese@GMX.DE> wrote:

> Hi all,
>
> I have written a macro which runs logistic
> regressions (PROC LOGISTIC). Now I would like to
> write the odds ratio and the lower and upper
> confidence limit to a dataset for each run of the
> macro.
> I read somewhere that the confidence limits can't be
> written to a dataset when using the STRATA option.
> Is that right and is there nevertheless a way to do
> it?
>
> Kind regards,
> N. Knese
> --
> "Feel free" - 5 GB Mailbox, 50 FreeSMS/Monat ...
> Jetzt GMX ProMail testen:
> www.gmx.net/de/go/mailfooter/promail-out
>


Cheers!
Albert-Jan

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Did you know that 87.166253% of all statistics claim a precision of results that is not justified by the method employed? [HELMUT RICHTER]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



____________________________________________________________________________________
Now that's room service!  Choose from over 150,000 hotels
in 45,000 destinations on Yahoo! Travel to find your fit.
http://farechase.yahoo.com/promo-generic-14795097