BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:MRC Biostatistics Unit Seminars
SUMMARY:“Extended multivariate generalised linear and non-
linear mixed effect models” - Dr Michael Crowther
\, Biostatistics Research Group\, Department of He
alth Sciences\, University of Leicester
DTSTART;TZID=Europe/London:20171005T143000
DTEND;TZID=Europe/London:20171005T153000
UID:TALK81621AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/81621
DESCRIPTION:Multivariate data occurs in a wide range of fields
\, with ever more flexible model specifications be
ing proposed\, often within a multivariate general
ised linear mixed effects (MGLME) framework. In th
is talk\, I’ll describe some current work developi
ng an extended framework\, encompassing multiple o
utcomes of any type\, each of which could be repea
tedly measured (longitudinal)\, with any number of
levels\, and with any number of random effects at
each level. Many standard distributions are descr
ibed\, as well as non-standard user-defined non-li
near models. The extension focuses on a complex li
near predictor for each outcome model\, allowing s
haring and linking between outcome models in an ex
tremely flexibly way\, either by linking random ef
fects directly\, or the expected value of one outc
ome (or function of it) within the linear predicto
r of another. Non-linear and time-dependent effect
s are also seamlessly incorporated to the linear p
redictor through the use of splines or fractional
polynomials. I’ll further discuss level-specific r
andom effects distributions and numerical integrat
ion techniques to improve usability\, relaxing the
normally distributed random effects assumption to
allow multivariate t-distributed random effects.
I’ll consider some special cases of the general fr
amework\, describing some new models in the fields
of clustered survival data\, joint longitudinal-s
urvival models\, and discuss various potential use
s of the implementation. User friendly\, and easil
y extendable\, software is provided.
LOCATION:Large Seminar Room\, 1st Floor\, Institute of Publ
ic Health\, University Forvie Site\, Robinson Way\
, Cambridge
CONTACT:Alison Quenault
END:VEVENT
END:VCALENDAR