Emulators for Computer Simulation Models, 10am-4.30pm, Monday 13
September 2010 at the Holiday Inn in Brighton
Presented by Tony O'Hagan, Peter Challenor and other members of the MUCM
Project (University of Sheffield) this course will provide an
introduction to the use of emulators to make inference about the
behaviour of complex computer simulation models based on a limited
number of model runs. An emulator is a statistical representation of the
simulator, providing estimates of what the simulator output would be at
untried inputs, together with quantified uncertainty. Having built an
emulator, we can not only predict simulator outputs but also efficiently
address questions such as uncertainty analysis, sensitivity analysis and
calibration of models.
The course is primarily practical. There will be one session on
principles and ideas, another on the MUCM toolkit as a resource for
building and using emulators, and the remainder will be case studies.
Attendees will learn about an emerging area of serious statistical and
scientific interest. They will gain appreciation of the power of
emulator methods and what is involved in implementing them.
The course is aimed at statisticians working in fields where complex
models are used for prediction, understanding, optimisation or policy
making. This could be almost any area of science, technology, industry
or government.
Some appreciation of the Bayesian approach would be helpful to
attendees.
The course forms part of the pre-conference courses and workshops
programme at the RSS 2010 Conference in Brighton - information about
registration can be found here: http://www.rss.org.uk/main.asp?page=3192
Paul Gentry
Theme Manager, Meetings & Conferences
Royal Statistical Society
Direct Tel. (020) 7614 3918
Fax. (020) 7614 3905
Register online for RSS 2010 Conference: www.rss.org.uk/rss2010
______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|