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


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 <>
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 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
phone +46 18 471 4105
fax +46 18 471 4003