Joint Seminar: Assessing model discrepancy using a multi-model ensemble

Any model-based prediction must take account of the discrepancy
between the model and the underlying system.  In physical systems such
as climate, where a typical system component is indexed by space,
time, and type, this discrepancy has a complex joint structure, which
makes direct elicitation very demanding.  Here we propose an
alternative to direct elicitation, based on judgements about a
collection of model-evaluations, known as a Multi-Model Ensemble
(MME). The crucial statistical modelling framework is that of
second-order exchangeability, within a Bayes linear treatment.  We
show how a second-order exchangeable MME can be used to learn about
the discrepancy, and also how it can be used to support our judgements
about the relation between the model-evaluations and the system.  We
illustrate our approach with global surface temperature, using an MME
constructed for the IPCC Fourth Assessment Report.

These are both from a paper that we have just finished, available at




15:15 Uhr


Bundesstr. 53, room 022/023
Seminar Room 022/023, Ground Floor, Bundesstrasse 53, 20146 Hamburg, Hamburg


Jonathan Rougier, University of Bristol


Zur Übersicht