University of Edinburgh
School of Mathematics

Statistics seminars, Spring 2010

Friday 12 February, 3:15pm, JCMB 5327 - Glenn Marion (Biomathematics  
and Statistics Scotland)
"Inference in continuous time discrete-state stochastic models:  
applications in
ecology, epidemiology and animal behaviour"

Friday 26 February, 3:15pm, JCMB 5327 - Karim Lounici (University of  
Cambridge)
"Sparsity oracle inequalities in high-dimensional statistical problems"

Friday 12 March, 3:15pm, JCMB 5327 - David Signorini (Scottish Government)
"Statistical challenges in Justice: How does evidence influence policy?"

Monday 12 April, 2pm, JCMB 5215 - N. Balakrishnan (McMaster University)
Title: tbc

--------------------------------------------------------
Abstracts
--------------------------------------------------------

Friday 12 February, 3:15pm, JCMB 5327

Glenn Marion (Biomathematics and Statistics Scotland)

"Inference in continuous time discrete-state stochastic models:  
applications in
ecology, epidemiology and animal behaviour"

Abstract
Discrete state-space Markov processes provide a remarkably flexible  
framework both to describe and infer the behaviour of a broad range of  
systems. I will outline how to conduct statistically sound parameter  
estimation for such models when, as is typically the case, only  
partial observations are available. Key issues and ongoing problems  
associated with this approach will be illustrated via examples taken  
from recent work such as: agent-based modelling of ruminant behaviour;  
the spread of alien plants across the UK; and models of disease spread  
in experimental crop mixtures. If time permits I will also discuss  
ongoing work on the applicability of Rissanen's minimum description
length principle to model selection for such processes.

--------------------------------------------------------
Friday 26 February, 3:15pm, JCMB 5327

Karim Lounici (University of Cambridge)

"Sparsity oracle inequalities in high-dimensional statistical problems"

Abstract.
This talk is about statistical learning in high-dimension, that is  
when the number of parameters to estimate is larger than the sample  
size. In this context, the generally adopted assumption is that the  
number of active parameters is much smaller than the number of  
potential parameters. This assumption is called the "sparsity  
assumption". We study the statistical properties of two types of  
procedures: penalized risk minimization
procedures with l_1-type penalty such as the Lasso or the group Lasso  
and exponential weights procedures.

--------------------------------------------------------
Friday 12 March, 3:15pm, JCMB 5327

David Signorini (Scottish Government)

"Statistical challenges in Justice: How does evidence
influence policy?"

Abstract
Evidence-based policy is now well established within
Government. But what does this mean in practice? This talk will explore
some of the analytical challenges and problems that have come up in
Justice over the last few years, and reflect on how statistics and
statisticians can and should be used to help deliver Government policy.

--------------------------------------------------------
Monday 12 April, 2pm, JCMB 5215

N. Balakrishnan (McMaster University)

Title and abstract: tbc


-- 
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.