KEELE UNIVERSITY
CENTRE FOR MEDICAL STATISTICS
SEMINAR SERIES on BIOSTATISTICS 2002/2003
NO 3: Wednesday, December 11th 2002, at 2:30 pm
"Longitudinal analysis of repeated binary data using autoregressive and
random effect modelling"
Professor Murray Aitkin
University of Newcastle
This talk discusses an extension of random effect models for
longitudinal binary data to allow for the assessment of both serial
(AR1) dependence and random effect structure, by combining
autoregressive and random effect modelling.
Conditioning on the first outcome involves serious loss of information
about the random effect variance component, as well as losing
information about this outcome.
An extended interaction model allows for full information on both the
first and the subsequent outcomes. This requires only the random effect
model fitting procedures. Longitudinal binary data on childhood obesity
are used to illustrate the procedure.
Murray Aitkin, University of Newcastle
and
Marco Alfo', Universita La Sapienza, Rome
All welcome!!
Venue:
Room 2.22, Third Floor
MacKay Building
Keele University
http://www.keele.ac.uk
http://www.keele.ac.uk/depts/ma/seminars/medstats.html
http://www.keele.ac.uk/university/campus/maps/
*****************************
Janet Drewery
Centre for Medical Statistics
MacKay Building
Keele University
Keele, Staffs ST5 5BG
Tel: (01782) 583269
Fax: (01782) 583269/584268
E-mail: [log in to unmask]
*****************************
|