BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Uncertainty Quantification - 
DTSTART:20200702T083000Z
DTEND:20200702T103000Z
UID:TALK149791@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:<b>Chair: <br></b><span>Peter Challenor   <b>Why do uncertaint
 y quantification</b>  </span><br><b>Speakers\;</b><br><span>Evan Baker (Ex
 eter<b>) </b><span><b>Emulating Stochastic Models</b><span><b> </b><br></s
 pan></span></span>  Building emulators for complex models typically involv
 es Gaussian processes. For stochastic models\, the flexibility of a Gaussi
 an process is a nice feature\, but modifications are needed to account for
  the noisiness of simulations. In this talk I will summarise some key attr
 ibutes of stochastic models and how these can change the emulation methodo
 logy. Additionally\, I will briefly talk about the simulation design issue
 s that arise for stochastic models.  <br>Jeremy Oakley (Sheffield)&nbsp\; 
 <span><span>-<b> Introduction to Probabilistic Sensitivity Analysis </b><b
 r> Mathematical models of infectious diseases invariably have uncertainty 
 about the correct values of some of their model inputs/parameters. This in
 duces uncertainty in the model outputs. In some situations\, it may be des
 irable to reduce this uncertainty\, by collecting more data about uncertai
 n model inputs\, before using the model outputs to inform decisions. Howev
 er\, it is unlikely that all inputs are &#39\;equally important&#39\;: som
 e will contribute to output uncertainty more than others. I will discuss h
 ow probabilistic sensitivity analysis can be used to identify which uncert
 ain inputs are most influential\, and describe simple computational tools 
 that can be used for implementing the analysis\, based on a random sample 
 of model runs.</span>  <br></span><br><br><br><br><br><br><br>
LOCATION:Seminar Room 2\, Newton Institute
END:VEVENT
END:VCALENDAR