A reminder of our meeting in two days time:
THE CAMBRIDGE STATISTICS DISCUSSION GROUP
Thursday 4th May 2006 7:15 for 7:45
Hopkinson Room,
University of Cambridge Computer Laboratory,
Pembroke Street, Cambridge
Parameter estimation and prediction for the course of a
single epidemic
Adam Kleczkowski,
Department of Plant Sciences,
Cambridge
Abstract:
Predictions in epidemiology are mainly made on the basis of a mean behaviour
of an ensemble of replicates. However, in practical applications we are
often interested in predicting a course of a disease in a single particular
population. Close inspection of many examples of replicate disease progress
curves shows that once a trajectory is chosen, the variability is low. In
contrast, the between-replicate variability is often very high, resulting in
a high degree of unpredictability based on the average behaviour. A
prediction of a future course of an epidemic can be based on partial
observations of this population and on complete past observations of similar
populations. A hierarchical Bayesian approach is a natural choice for a
replicate-based prediction method, enabling us to combine partial
information about a particular realisation with a behaviour of other
realisations. My talk explores this idea for a set of contrasting population
dynamics of a plant pathogen. We show that even a very limited set of data
can be used to narrow down predictions for development of a single epidemic,
even if the variability between individual epidemics is very large.
This has profound implications for data collection and analysis as well as for
a design of control strategies. We also discuss the relevance of our results
in a broader ecological context, particularly for describing invasion and
persistence of species.
Speaker: Adam did his postgraduate study at the Institute of Physics in Krakow
fitting deterministic and stochastic models to physical processes and,
subsequently in Germany, to ecological systems. Since coming to Cambridge
he has worked in various university departments modelling behaviour of
measles, soil-born diseases and cereal pathogens.
He is currently lecturing in Mathematical Biology at the Department of Plant
Sciences looking at plant and microbial systems biology.
Directions:
Directions: Enter the New Museums site through the arch from Pembroke Street.
Turn to the left in front of the building with the brown iron staircase
and pass through the passage way by the tree. Enter the lobby which is
straight in front of you and find the back entrance to the Computer Service
on the left. When inside follow the signs to the Hopkinson Room.
It is not possible to park on the New Museums site in the evening.
The back entrance to the Computer Service will be locked at 7:45.
Arrivals after 7:45 can gain admittance by contacting the secretary on
07761769436.
Provisional Next Meeting:
3rd October - Nick Galwey (GSK).
Supper: Some members eat regularly in the University Centre before
each meeting at 6pm. Feel free to join them.
Subscriptions: of 4 pounds are now due for attending the 2005-2006 session.
Secretary: Peter Watson, MRC Cognition and Brain Sciences Unit,
15 Chaucer Road, Cambridge CB2 2EF;telephone 01223 355294 Extension 801;
E-mail [log in to unmask]
