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 (June 2003, week 3)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Wed, 18 Jun 2003 14:21:49 -0800
Reply-To:     Dan Jeffers <ojodelmar@HOTMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Dan Jeffers <ojodelmar@HOTMAIL.COM>
Subject:      binary outcomes and repeated observations
Content-Type: text/plain; format=flowed

Dear SAS-L readers,

I am working with a dataset composed of a series of explanatory variables and a binary response. My approach to analysis had been using PROC GENMOD to model the probability of the occurrence of the response in relation to the explanatory variables. However, the data are from repeated observations taken from the same set of subjects over time.

Is there a maximum likelihood based approach to analyzing this type of data? I have read of population-average logistic regression models to deal with correlation within subects, is this type of analysis available in SAS? ALso, is there a way to look at the correlation within observations taken on the same subjects to see if the correlation is significant to potentially justify taking a standard logistic regression approach as if all observations were independent? And finally, is the problem of correlation among outcomes measured from the same subject related to the problem over overdispersion, i.e. will acounting for overdispersion using a variance inflation factor deal with correlation among repeated observations?

Thanks for your help.

Dan Jeffers

_________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=features/junkmail


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