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 (May 2008, week 1)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Thu, 1 May 2008 17:45:50 -0400
Reply-To:     Peter Flom <peterflomconsulting@mindspring.com>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Peter Flom <peterflomconsulting@MINDSPRING.COM>
Subject:      Re: Hierarchical Models--Centering predictors--WHY
Comments: To: Tom White <tw2@MAIL.COM>
Content-Type: text/plain; charset=UTF-8

Tom White <tw2@MAIL.COM> wrote

>Thnaks Peter, Actually, I'm looking for a book on multi-level logistic >regression becasuse my data are nested(health care claims on patients who >are nested inside doctors) and the dependent variable is binomial(FRAUD=1(yes) >or FRAUD=0(no)). So, I need to be able to correctly use PROC GLIMMIX in >SAS to fit this type of data. Many of the books I see talk about a whole >bunch of other models that aren't logistic in nature. But I still don't >know of any good book to show me in SAS all the options about fitting a >logistic curve to my very skewd data (99% 0s=no-fraud and 1% 1s=fraud). >So, there are many issues to deal with binomial type of data and I don't >wish to study right now other modelsbeyond multi-level logist with PROC >GLIMMIX. Thnaks. T

OIC..... sorry, I was confused as to what you wanted.

For this, there is one chapter in Littel et al. on GLIMMIX.

There are also some good SGF and SUGI presentations by Oliver Schabenberger.

From what I know, highly skewed data is not really a problem, as long as the sample size in the small group is large enough

Peter

Peter L. Flom, PhD Statistical Consultant www DOT peterflom DOT com


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