Date:         Tue, 29 Oct 2002 16:42:12 -0800
Reply-To:     Jonathan Greenberg <greenberg@UCDAVIS.EDU>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Jonathan Greenberg <greenberg@UCDAVIS.EDU>
Subject:      Longitudinal data analysis question?
Content-Type: text/plain; charset="US-ASCII"

Hi there, I'm trying to analyze longitudinal data, with a known model:

I have 61 observations, 4 variables per observations, and they were recorded
on 5 unevenly spaced dates.  The model should be exponential, so for each
variable:

Variable[t] = Variable[0] * e ^ (k * t)

I want to be able to estimate Variable[0] and k for each observation, or, if
possible, derive a single model assuming the rates, k, for all observations
are constant, but the starting times, Variable[0], might be different.

I think PROC MIXED might be able to do this, but am a SAS newbie.  Any
suggestions?  I want to know, for any variable value, what is the time
elapsed since the initial time?

--j