s-news
[Top] [All Lists]

Mixture modeling using S-Plus?

To: s-news@wubios.wustl.edu
Subject: Mixture modeling using S-Plus?
From: Savitri Appana <saviappana@yahoo.com>
Date: Tue, 23 Nov 2004 13:20:28 -0800 (PST)
Comment: DomainKeys? See http://antispam.yahoo.com/domainkeys
Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; b=aYaf6K70/MZOUBrrM3XadjDpitD6XyIbaEmIEkSatihZxOBJx1L0S/hCEgdN+2tXU0hcnCKVpLlHBfsJxgxDKd5H+F8IHJ3GTJyG6AVKRL6WY/r6FaMvXjuZlC5LlSFanYlVfPQi5pKr/2t7KNmK2ZVDkRit2FAtp0Oz1NzWghE= ;
Hello viewers,

(First of all, sorry for such a long message.)

I was recently asked to model some growth curve data
using NLME.  I've used the self-start function,
SSasymp(Asym, RO, lrc, time), and fit the groupedData
accordingly.  Once I ran the model, i used augPred to
generate the prediction plots to see the fits.  For
most animals the fits were nice with clearly defined
asymptotes, etc... however, some animals kept on
increasing  in weights (due to, we think, menopause). 


Now i'm asked to look into a model of mixture of two
asymptotic curves (one of which involves time delay). 
Any suggestions on how i can go about doing this is
much appreciated.

I've looked (searched online, journals, etc.) and was
unsuccessful in finding literature dealing with this
topic.  So if someone came across any references on
this topic, i hope you can point me in that direction.

Thanks in advance,
Savi

Statistician, FDA/NCTR
email: saviappana@yahoo.com



__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 

<Prev in Thread] Current Thread [Next in Thread>
  • Mixture modeling using S-Plus?, Savitri Appana <=