event
Thursday 06 Oct 2016: Statistical Science Seminar: Ensemble averaging and mean squared error
Jonty Rougier - University of Bristol
H103 14:30-16:30
In fields such as climate science, it is common to compile an ensemble
of different simulators for the same underlying process. It is a
striking observation that the ensemble mean often out-performs at least
half of the ensemble members in mean squared error (measured with
respect to observations). In fact, as demonstrated in the most recent
IPCC report, the ensemble mean often out-performs all or almost all of
the ensemble members across a range of climate variables. This paper
shows that these could be mathematical results based on convexity and
averaging, but with implications for the properties of the current
generation of climate simulators.