From: "Mike Cole" <Michael.Cole@newcastle.ac.uk>
Subject: Help please...
Date: Wed, 13 Dec 2000 10:52:01 +0000

Hello

Any help or advise would be much appreciated. I am using a straightforward 2 compartment model to obtain 'empirical bayes' (posthoc) estimates of clearance for new patients based upon 1 or 2 data points for that individual together with data from 60 patients for which we have 6/7 data points per individual.

I would like to produce a graph showing the observed data for the new individual together with predictions based upon the 'empirical bayes' estimates for that patient. If I append the data file with dummy time points which have missing dependent variables the estimates become nonsensical.

Can anyone explain why this happens or suggest a simple alternative.

Thanks in advance.

Mike Cole
________________________________________________
Michael Cole
Dept of Statistics/Cancer Research Unit,
Merz Court, University of Newcastle upon Tyne,
Newcastle upon Tyne, NE1 7RU, UK
Tel: 0191 222 7209 (Stats)/ 0191 222 7134 (CRU)
Fax: 0191 222 8020
________________________________________________


*****


From: Mats Karlsson <Mats.Karlsson@biof.uu.se>
Subject: Re: Help please...
Date: Wed, 13 Dec 2000 12:15:27 +0100

Hi Mike,

If you create an EVID column in your data file and give EVID the value of:
0 for true observations
1 for dose
2 for dummy points,

you should obtain what you want.

Best regards,
Mats
--
Mats Karlsson, PhD
Professor of Biopharmaceutics and Pharmacokinetics
Div. of Biopharmaceutics and Pharmacokinetics
Dept of Pharmacy
Faculty of Pharmacy
Uppsala University
Box 580
SE-751 23 Uppsala
Sweden
phone +46 18 471 4105
fax +46 18 471 4003
mats.karlsson@biof.uu.se