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:Uncertainty Quantification -
DTSTART;TZID=Europe/London:20200702T093000
DTEND;TZID=Europe/London:20200702T113000
UID:TALK149791AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/149791
DESCRIPTION:**Chair: **

Peter Challenor **Why d
o uncertainty quantification**

**Sp
eakers\;**

Evan Baker (Exeter**) **<
span>**Emulating Stochastic Models**** <
/b>**

** Building emulators f
or complex models typically involves Gaussian proc
esses. For stochastic models\, the flexibility of
a Gaussian process is a nice feature\, but modific
ations are needed to account for the noisiness of
simulations. In this talk I will summarise some ke
y attributes of stochastic models and how these ca
n change the emulation methodology. Additionally\,
I will briefly talk about the simulation design i
ssues that arise for stochastic models. **

Jerem
y Oakley (Sheffield) \; -** Intro
duction to Probabilistic Sensitivity Analysis **

Mathematical models of infectious diseases in
variably have uncertainty about the correct values
of some of their model inputs/parameters. This in
duces uncertainty in the model outputs. In some si
tuations\, it may be desirable to reduce this unce
rtainty\, by collecting more data about uncertain
model inputs\, before using the model outputs to i
nform decisions. However\, it is unlikely that all
inputs are '\;equally important'\;: some wi
ll contribute to output uncertainty more than othe
rs. I will discuss how probabilistic sensitivity a
nalysis can be used to identify which uncertain in
puts are most influential\, and describe simple co
mputational tools that can be used for implementin
g the analysis\, based on a random sample of model
runs.

LOCATION:Seminar Room 2\, Newton Institute
CONTACT:INI IT
END:VEVENT
END:VCALENDAR