LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (August 1998, week 2)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Tue, 11 Aug 1998 13:23:59 -0500
Reply-To:     gkeeler@PSYCH.MC.DUKE.EDU
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         gkeeler@PSYCH.MC.DUKE.EDU
Organization: Duke University, Durham, NC, USA
Subject:      Re: Weighted Logistic Regression
Content-Type: text/plain; charset=us-ascii

mt_rogel@my-dejanews.com wrote:

> The sampling design used was a two-stage stratified cluster and i would like > to > incorporate this when i use logistic regression. how i can do this.

Trisha, PROC LOGISTIC by itself does NOT incorporate the weight variable correctly. You can check the results of the procedure yourself by running it with and without the weight statement. I seem to remember that the calculated variance doesn't change when you add the weight statement.

There are several ways to get around this error.

1) The LOGITSE macro adjusts the variance using the "sandwich" correction factor. It is available at the SAS website but I will mail a copy of it directly to you.

2) Use the GLIMMIX macro for logistic regression using the EMPIRICLE option. This macro is installed with SAS in the STAT\SAMPLE directory.

3) Use PROC GENMOD.

Good Luck...........Gordon


Back to: Top of message | Previous page | Main SAS-L page