University of Edinburgh
School of Mathematics and BioSS
Statistics seminar
Friday 28 March, 3:10pm, JCMB5323 - Amy Wilson (University of
Edinburgh) "The Evaluation of Evidence Relating to Traces of Drugs on
Banknotes"
Thursday 3 April, 3:10pm, JCMB5327 - Marjan Sjerps (University of
Amsterdam) "How strong is the evidence?"
There will also be a School of Mathematics Colloquium event:
Friday 4 April 4:10pm (JCMB5215) Valerie Isham (University College
London) "Stochastic modelling in hydrology"
Abstracts are given below.
This seminar is a part of Maxwell Institute seminar series.
Website: http://www.maths.ed.ac.uk/events/statistics
----------------
Amy Wilson (University of Edinburgh) "The Evaluation of Evidence
Relating to Traces of Drugs on Banknotes"
Abstract: Much research in recent years for evidence evaluation in
forensic science has focussed on methods for determining the
likelihood ratio for different evidential data types. The likelihood
associated with a set of evidential data is calculated under each of
two propositions, that proposed by the prosecution and that proposed
by the defence. The value of the evidence is given by the ratio of the
likelihoods of these two propositions. The aim of this research is to
evaluate this likelihood ratio under two scenarios. The first is when
the evidence consists of continuous autocorrelated data. The second,
an extension to this, is when the observed data are also believed to
be driven by an underlying latent Markov chain. Four models have been
developed to take these attributes into account: an autoregressive
model of order one, a hidden Markov model with autocorrelation between
adjacent data points and a nonparametric model with two different
bandwidth selection methods.
Application of these methods will be illustrated with an example where
the data relate to traces of cocaine on banknotes. The prosecution
proposition for this example is that the banknotes are associated with
a person who is involved with crime involving cocaine. Conversely, the
defence proposition is that the banknotes are associated with a person
who is not involved with crime involving cocaine. The likelihood ratio
associated with these two propositions will be evaluated using each of
the four models, and the results compared.
Marjan Sjerps (University of Amsterdam) "How strong is the evidence?"
Abstract: Forensic statistics is concerned with questions like: how
can we evaluate the strength of forensic evidence like DNA profiles or
shoe
prints, how can we derive the combined evidential strength and how can
we communicate this to a lay judicial decision maker (a judge or jury
member)? In this talk I will focus on various topics that caused some
debate, such as the way to combine evidence and the way that sampling
uncertainty affects the evidential value. I will present these topics in
the context of shoeprint evidence in a real criminal case that took
place in the Netherlands.
Valerie Isham (University College London) "Stochastic modelling in hydrology"
Abstract: Rainfall is the driving force for many hydrological
processes. As has been all too apparent in recent months, rainfall
that cannot be absorbed or drained away causes major flooding
disasters and flood defences must be designed to cope with extreme
events. Soil moisture, for which rainfall is the input, provides the
dynamic link between climate, soil and vegetation, and impacts plant
dynamics as well as other processes at a range of spatial scales.
Historical rainfall data are, perhaps surprisingly, often not
available at the temporal and spatial resolution needed for
hydrological modelling and design. The talk will describe some
stochastic models for rainfall fields in continuous space-time that
can be used to provide artificial rainfall data at arbitrary temporal
and spatial scales, together with some simple soil moisture models for
which these rainfall models form the input. Climate change poses an
additional challenge, as rainfall under future climate scenarios is
needed for hydrological design, and a way to allow for this in model
fitting will be described.
--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|