=======================================================================
SEMINARS at the MRC BIOSTATISTICS UNIT, Cambridge
LENT TERM 2004
=======================================================================
27th January
Dr. M. Plummer,
International Agency for Research on Cancer,
Lyon, France.
Title: Bayesian hierarchical modelling with JAGS.
Abstract:
JAGS is Just Another Gibbs Sampler. It is a program for the analysis of
Bayesian graphical models using Gibbs Sampling, closely based on the
program BUGS.
There are two primary motivations for JAGS. The first is to have an
extensible BUGS "engine", allowing users to analyze non-standard
problems without having to write their own software from scratch. This
will be illustrated using multi-state Markov models for disease
progression. A second motivation is to provide a platform for reference
implementations of methods in Bayesian modelling, and so encourage the
widespread dissemination of these methods. This will be illustrated by
some of my own ideas on Bayesian model criticism.
=======================================================================
10th February
Professor A.P. Dawid,
Department of Statistical Science, University College London.
Title: Bayes's Theorem And Weighing Evidence by Juries.
Abstract:
Although at first sight there may appear to be little connection
between Statistics and Law, the problems they tackle are in many ways
identical: each is concerned with interpretation of evidence.
However, lawyers and juries are (often proudly) ignorant of the
logical principles and methods that statisticians have developed to
tackle this, which can lead to serious misinterpretations of
statistical evidence. I discuss the uses and misuses of statistical
arguments in Court, emphasising in particular the potential of
likelihood and Bayesian approaches to shed light on subtle issues.
This is illustrated by my experiences as an expert witness for the
defence in the first (unsuccessful) appeal by Sally Clark against her
conviction for the murder of her two babies.
=======================================================================
9th March
Professor C. Glasbey,
Biomathematics & Statistics Scotland,
University of Edinburgh.
Title: How to segment 3-D images and analyse 1-D electrophoresis gels.
Abstract:
In this talk, the elegant method of Dynamic Programming (DP) will be
introduced in a non-technical way, and extensions will be considered for
when DP is not immediately applicable. DP is a
computationally-efficient method for finding the global solution to some
optimisation problems. For example, it can be used to track boundaries
in order to automatically segment 2-D medical images into different
anatomical regions (Glasbey and Young, 2002). It can also be used to
align pairs of tracks in 1-D electrophoresis gels, using the method of
Dynamic Time Warping which is also used in automatic speech recognition.
However, if images are three dimensional, or many gel tracks need
aligning, then simple DP is not possible. Extensions to DP will be
considered, illustrated by applications in 3-D X-ray computed tomography
and pulsed field gel electrophoresis.
Glasbey, C.A. and Young, M.J. (2002). Maximum a posteriori estimation
of image boundaries by dynamic programming. Applied Statistics, 51,
209-221.
=======================================================================
18th March
Professor A. R. Willan,
Population Health Sciences, Hospital for Sick Children
and Public Health Sciences, University of Toronto.
Title: The value of information and optimal clinical trial design.
Abstract:
Traditional sample size calculations for randomized clinical trials depend
on arbitrarily chosen factors, such as type I and II errors. Type I error,
the probability of rejecting the null hypothesis of no difference when it
is true, is most often set to 0.05, regardless of the cost of such an error
In addition, the traditional use of 0.2 for the type II error means that the
money and effort spent on the trial will be wasted 20% of the time even when
the true treatment difference is equal to the smallest clinically important
one and, again, will not reflect the cost of making such an error. A pragmatic
trial (otherwise know as an effectiveness trial or management trial) is
essentially an effort to inform decision-making, i.e. should Treatment be
adopted over Standard? As such, a decision theoretic approach will lead to
a more optimal sample size determination. Using incremental net benefit and
the theory of the expected value of information, and taking a societal
perspective, it is show how to determine the sample size that maximizes
the difference between the cost of doing the trial and the value of the
information gained from the results. The methods are illustrated using
examples from oncology and obstetrics.
=======================================================================
Seminars start at 2.30pm in the Large Seminar Room, 1st Floor,
Institute of Public Health, University Forvie Site, Robinson Way,Cambridge.
All welcome to attend.
=======================================================================
Daniela De Angelis
MRC Biostatistics Unit Tel: + (0)1223 330390
Institute of Public Health
Robinson Way Fax: + (0)1223 330388
Cambridge
CB2 2SR Web: www.mrc-bsu.cam.ac.uk
United Kingdom
|