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:Isaac Newton Institute Seminar Series
SUMMARY:A Comparison of Approximate Bayesian Computation a
nd Stochastic Calibration for Spatio-Temporal Mode
ls of High-Frequency Rainfall Patterns - Matthew P
ratola (Ohio State University)
DTSTART;TZID=Europe/London:20180412T153000
DTEND;TZID=Europe/London:20180412T160000
UID:TALK103738AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/103738
DESCRIPTION:Modeling complex environmental phenomena such as r
ainfall patterns has proven challenging due to the
difficulty in capturing heavy-tailed behavior\, s
uch as extreme weather\, in a meaningful way. Rec
ently\, a novel approach to this task has taken th
e form of so-called stochastic weather generators\
, which use statistical formulations to emulate th
e distributional patterns of an environmental proc
ess. However\, while sampling from such models is
usually feasible\, they typically do not possess
closed-form likelihood functions\, rendering the u
sual approaches to model fitting infeasible. Furt
hermore\, some of these stochastic weather generat
ors are now becoming so complex that even simulati
ng from them can be computationally expensive. We
propose and compare two approaches to fitting com
putationally expensive stochastic weather generato
rs motivated by Approximate Bayesian Computation a
nd Stochastic Simulator Calibration methodologies.
The methods are then demonstrated by estimating
important parameters of a recent stochastic weathe
r generator model applied to rainfall data from th
e continental USA.
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:INI IT
END:VEVENT
END:VCALENDAR