Premium
A note on parameter estimation for variant Creutzfeldt–Jakob disease epidemic models
Author(s) -
Clarke Paul S.,
Ghani Azra C.
Publication year - 2006
Publication title -
statistics in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.996
H-Index - 183
eISSN - 1097-0258
pISSN - 0277-6715
DOI - 10.1002/sim.2567
Subject(s) - estimation , maximum likelihood , computer science , incubation period , statistics , econometrics , scale (ratio) , mathematics , geography , biology , cartography , incubation , economics , management , biochemistry
A recent series of papers has raised issues regarding estimation of the key epidemiological parameters of variant Creutzfeldt–Jakob disease (vCJD) from fitting survival models to case data. In particular, it was stated that the scale of the epidemic cannot be estimated and must be fixed in any analysis. We show that this problem is an artefact of the approximate likelihood used in these papers to facilitate model‐fitting, and is not a concern if estimation is based on the full likelihood. We also address comments in these papers regarding the choice of incubation period distribution. Copyright © 2006 John Wiley & Sons, Ltd.