s-news
[Top] [All Lists]

Random effects model for time series data

To: s-news@lists.biostat.wustl.edu
Subject: Random effects model for time series data
From: Martyn Colins <martyncollins10@yahoo.com>
Date: Thu, 7 Sep 2006 14:00:54 -0700 (PDT)
Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=Message-ID:Received:Date:From:Subject:To:MIME-Version:Content-Type:Content-Transfer-Encoding; b=wwc2qqe5PqoFV2eyy+RylweexNJjf27CTOaF8SAvcQwQ6UIi8FMgN4gEKwwbQEuXvoyK8D6BwMDFeCk1Vc07+PCJCpQfG4ll6uhGaptLFxc/SYTtXMXtwi3WUu5vjWJKuImDN2Fg42R77Hlc6Spt7CTwVFCcyNjrfdYTgrYxQt4= ;
 
I would like to know how to fit a random effects model to equally spaced time series data. The data is multivariate normal data, I have 11,000 subjects each with 72 monthly observations spanning 6 years. I have 3 fixed effects and 1 random effect, these effects are the same for all 72 observations (i.e. the same model for each response).
 
I'm would like to know how to fit such a model in S-Plus?
 
Also, I'm interested in finding out the correlations between the fixed effects and the random effect? I would be very grateful for your suggestions.
 
Regards,
Martyn
 
 


All-new Yahoo! Mail - Fire up a more powerful email and get things done faster.
<Prev in Thread] Current Thread [Next in Thread>