UNCERTAINTY IN SIMULATION MODELS WORKSHOP

10th July 2009, National Oceanography Centre Southampton

 

The presentations are available through the following links:

Tony O'Hagan, University of Sheffield	Introduction: Simulators & Emulators
Jeremy Oakley, University of Sheffield	Case Study - Sensitivity Analysis in Epidemiological Modelling
Dan Cornford, Aston University	How to Emulate: Recipes without Patronising (the MUCM Toolkit)
Peter Challenor, NOCS	Estimating Risk from a Climate Model
Ian Vernon, University of Durham	Calibrating the Universe:a Bayesian Analysis of a Galaxy Formation Simulation

 

To read an article about the workshop by The Insitute of Mathematics and its Applications, please follow this link.

 

WORKSHOPS FOR MODELLERS, MODEL USERS AND DECISION-MAKERS

2nd & 3rd July 2008, Durham University

MUCM held two one-day workshops in July for those who build, use or rely on science-based models for prediction of complex real-world processes. These workshops focussed on applications in the biological and health sciences, although the MUCM methods are applicable much more generally.

 

Dissemination Workshop, 2nd July 2008

The first workshop explained the power of MUCM methods to quantify and analyse the uncertainties in the outputs of complex models. Tony O'Hagan opened the workshop with an overview of the MUCM methods for uncertainty analysis, sensitivity analysis, calibration and validation of models. This was followed by a series of short case studies exemplifying their use in a range of applications. The workshop continued with a review by Michael Goldstein of ongoing research in the field and the future potential of these methods.

 

Slides used in each of the talks can be accessed below.

 

Tony O'Hagan, University of Sheffield	MUCM - an Overview
Peter Challenor, National Oceanography Centre, Southampton	Analysing Cellular Reactions - Yeast Metabolysis
John Paul Gosling, University of Sheffield	Speeding up and Extending the Sensitivity Analysis of an Epidemiological Model
Ian Vernon, Durham University	Uncertainty Analysis for Systems Biology Models
Jonathon Rougier, University of Bristol	Predicting Snow Behaviour in Large Chute Experiments
Michael Goldstein, Durham University	General Issues in Uncertainty Analysis for Complex Models

 

Hands-on Workshop, 3rd July 2008

The second workshop provided a more detailed and practical introduction to some of the most basic MUCM methods through practical sessions using the GEM-SA and SHELF software packages.

 

Course Notes and links to software packages used in this workshop can be accessed below.

 

Course Notes	Tony O'Hagan, University of Sheffield
SHELF Software	The SHeffield ELicitation Framework (SHELF) is a package of documents, templates and software to carry out elicitation of probability distributions for uncertain quantities from a group of experts.
GEM-SA Software	The Gaussian Emulation Machine for Sensitivity Analysis (GEM-SA) will allow you to build an emulator of a computer code from a set of input and output points. It will also perform prediction, uncertainty analysis and sensitivity analysis of the code using far fewer code runs than Monte-Carlo based methods.