Date:         Thu, 9 Sep 2010 00:56:33 +0100
Reply-To:     John Whittington <John.W@MEDISCIENCE.CO.UK>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         John Whittington <John.W@MEDISCIENCE.CO.UK>
Subject:      Re: ESTIMATE statement in Proc MIXED
Comments: To: R B <ryan.andrew.black@GMAIL.COM>
In-Reply-To:  <AANLkTinh7VuTOiAaYbKDFYhYJe__k2rPs273KQ5e4i+r@mail.gmail.c om>
Content-Type: text/plain; charset="us-ascii"; format=flowed

At 16:57 08/09/2010 -0400, R B wrote:
>I don't remember if the OP stated whether or not the time intervals
>between observations were equal. If the time intervals between observations
>are quite different (e.g., 3 weeks between the first and second visit; 12
>weeks between the second and third visit), then he might be better off
>fitting a spatial covariance structure to capture the decay in correlations.
>
>I came across an article a while back that showed how to deal with attrition
>bias via the NLMIXED procedure.  While this approach may not be applicable
>to the OP's situation, others might find it useful.
>
>See link below:
>
>http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2733930/

Ryan, thanks for your interest.  The time intervals between repeat
observations in the data are all equal - although, as Dale has pointed out,
I may need to review my model in the light of the fact that I now have an
interest in 'adjacent time periods'.

I'll certainly look at the link you provided - it sounds potentially very
interesting, even if it does not prove to be applicable in my present
situation.

Kind Regards,


John

----------------------------------------------------------------
Dr John Whittington,       Voice:    +44 (0) 1296 730225
Mediscience Services       Fax:      +44 (0) 1296 738893
Twyford Manor, Twyford,    E-mail:   John.W@mediscience.co.uk
Buckingham  MK18 4EL, UK
----------------------------------------------------------------