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:Statistics
SUMMARY:Regret-regression for optimal dynamic treatment al
location\, without and with missing data - Robin H
enderson\, Newcastle University
DTSTART;TZID=Europe/London:20131129T160000
DTEND;TZID=Europe/London:20131129T170000
UID:TALK47612AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/47612
DESCRIPTION:A system is monitored over time. At time $t$ an in
put $A_t$ is determined and an output $S_{t+}$ obs
erved. The input is directly controllable by an
experimenter but the output $S_{t+}$ is not.\n \n
The statement above applies to a raft of problems
across multiple areas\, including control\, machin
e learning\, stochastic scheduling and many others
. A special case of growing interest in the biosta
tistical literature is the need for statistical m
ethodology aimed at determining optimal dynamic t
reatment regimes from observational data. Motivat
ion is primarily related to individualised medici
ne and causal inference.\n \nBuilding on the adva
ntage learning approach\, we propose a modelling
and estimation strategy that incorporates the regr
et functions of Murphy (2003) into a regression mo
del for observed responses. Estimation is quick an
d diagnostics are available\, meaning a variety of
candidate models can be compared. We consider th
ree issues relating to the missing data problem t
hat is ubiquitous in practical applications: estim
ation\, sub-optimal decisions and non-recovery. Th
e methods are illustrated using data on patients o
n long-term anticoagulation treatment.
LOCATION:MR12\, Centre for Mathematical Sciences\, Wilberf
orce Road\, Cambridge
CONTACT:
END:VEVENT
END:VCALENDAR