BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//ExpressionEngine Easy iCalendar plugin//NONSGML v1.1.1//EN
X-WR-TIMEZONE:Europe/Dublin
BEGIN:VEVENT
UID:14731
LOCATION:A2-002.
DTSTAMP:20170428T160010
DTSTART:20170428T160010
DTEND:20170428T170010
SUMMARY:Robust Joint Modelling: An introduction and exploration of r
ecent extensions
DESCRIPTION:MACSI\n\n \n\n The department of Mathematics and Statistics at t
he University of Limerick invites you to a seminar by Dr L
isa McCrink (Queen's University Belfast):\n\n Title: Robust
Joint Modelling: An introduction and exploration of recent e
xtensions\n\n Abstract: An introduction and exploration of rec
ent extensions With a steady growth in the volume of data be
ing collected currently\, there is an obvious need to build m
ore sophisticated tools with which to analyse this data. One
such method is joint modelling. In the last few decades\, jo
int modelling has proven to be a valuable technique in the s
imultaneous analysis of individuals’ repeated measurements
over time and the common link such longitudinal data has wi
th a time-to-event or survival process. This talk will discu
ss an extension to standard joint modelling techniques which
accounts for the presence of longitudinal outliers – the
use of robust joint models. In the joint modelling of longit
udinal and survival data it is commonly assumed that the ran
dom effects and error terms follow normality assumptions. Th
is research illustrates the negative impact of such conventi
ons when longitudinal outliers are present\, highlighting the
benefits of assuming that the random terms instead follow a
wider tailed t-distribution. The robust joint model present
ed is flexible as it allows the degree at which outliers are
adjusted for to be dictated by the data where the standard
joint model\, which assumes normality\, is given as a special
case. Such detrimental effects of outliers are clearly demon
strated within this research both through a simulation study
and an illustrative example focusing on Northern Irish rena
l patient data.\n\n \n\n This seminar will take place on Friday
\, April 28th \, at 4pm\, in Room A2-002.\n\n \n\n If you have an
y questions regarding this seminar\, please direct them to Ia
in Moyles (061 233726\, iain.moyles@ul.ie).\n\n A full list of
upcoming seminars can be found at http://www.ulsites.ul.ie/m
acsi/node/48011\n\n \n\n ---------------------------------------
-------------------------------------\n\n Supported by Science
Foundation Ireland funding\, MACSI - the Mathematics Applica
tions Consortium for Science and Industry (http://www.macsi.
ul.ie)\, centred at the University of Limerick\, is dedicated
to the mathematical modelling and solution of problems which
arise in science\, engineering and industry in Ireland.\n\n --
------------------------------------------------------------
--------------
END:VEVENT
END:VCALENDAR