Date:         Sun, 24 Jul 2005 15:19:40 -0400
Reply-To:     Kevin Roland Viel <kviel@EMORY.EDU>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Kevin Roland Viel <kviel@EMORY.EDU>
Subject:      Re: Appropriate subject-effects for REPEATED
In-Reply-To:  <70A9A1413B47DD4FB1F311A7C19B218C0521F8B6@m-nccd-1.nccd.cdc.gov>
Content-Type: TEXT/PLAIN; charset=US-ASCII

Matthew,

  I appreciate your response.  I do not have SAS/Genetics.  Half the point
of this exercise is to confirm my understanding of the statistical
analyses that I have performed in the other program.  In addition, SAS has
far superior data manipulation tools.

  At this point, the kinship coefficient encompasses all of the relevant
information concerning the relationship between two individuals.  I have
a listing of these coefficients for every distinct possible pair in a
given family.  All other pairs have 0 correlation (indepedent).  I have
constructed a matrix as a (series or) macro variable(s) to be used in the
TYPE = USER() option of the REPEATED statement.

  I will also explore the MIXED procedure.  I at least have the SI book
for that and my course notes from the SI class.  The extremely frustrating
aspect of my "career" to this point is sparse opportunity to apply these
tools, thus enforcing their mechanics and underlying assumptions.
Somehow, I do not think this would have given me (much) pause even two
years ago :(

Regards,

Kevin

Kevin Viel
Department of Epidemiology
Rollins School of Public Health
Emory University
Atlanta, GA 30322