Statistical Laboratory Seminars

Seminar Schedule:

This list is subject to revision

Select a date to view the relevant seminar abstracts:

Friday 26 January

2.00pm Pavel Gapeev (WIAS, Berlin)
Bayesian sequential testing and disorder problems for some diffusion processes

Friday 23 February

2.00pm Claire Hoggins (GlaxoSmithKline, on behalf of PSI (Statisticians in the Pharmaceutical Industry))
Careers for statisticians in the pharmaceutical industry

Friday 9 March

2.00pm Alexander J. McNeil (Heriot-Watt University)
Statistical Inference for Dependent Defaults and Credit Migrations

Abstracts and Further Details:

Friday 26 January

2.00pm Pavel Gapeev (WIAS, Berlin)

We study the Bayesian problem of sequential testing of two simple hypotheses about the local drift and the Bayesian problem of detecting a change in drift of an observed diffusion process. The optimal stopping times are found as the first times when the a posteriori probability processes leave the corresponding regions defined by stochastic boundaries depending on the observation process. It is shown that under some nontrivial relationships on the coefficients of the observed diffusion the problems admit closed form solutions. The method of proof is based on embedding the initial problem into two-dimensional optimal stopping problems and solving the equivalent free-boundary problems by means of the smooth-fit conditions.

Friday 23 February

2.00pm Claire Hoggins (GlaxoSmithKline, on behalf of PSI (Statisticiansin the Pharmaceutical Industry))

Claire Hoggins, a statistician working in the pharmaceutical industry, is visiting the university to talk about careers in the pharmaceutical industry The aim of the presentation is to raise awareness of the role of statistics within the drug development arena and to encourage students to consider a career as a statistician within the pharmaceutical industry. The talk will describe the pharmaceutical industry, give an overview of the drug development process and a detailed account of where a statistician or statistical programmer might be involved. Career opportunities and the working environments within the industry are to be presented and key skills and qualifications discussed. To conclude, an account of Claire.s career to date will be given, including an opportunity for the students to ask questions. Various PSI brochures (including a careers booklet written specifically for students considering a career in statistics) and contact details for any future questions will be provided.

Friday 9 March

2.00pm Alexander J. McNeil (Heriot-Watt University)

Any portfolio credit risk model that is to be used to calculate a loss distribution associated with defaults and changes in rating must address the challenge of modeling dependent defaults and dependent rating migrations. Most industry models (such as KMV, CreditMetrics, CreditRisk+) incorporate mechanisms for modeling this dependence, generally by assuming conditional independence of defaults and migrations given common economic factors. However, the calibration of these mechanisms is often quite ad hoc, despite the fact that the tail of the portfolio loss distribution is extremely sensitive to small changes in the parameters governing dependence.

We consider the problem of making formal statistical inference for such models based on historical default and rating migration data. In the solution we propose, portfolio credit models are represented as generalized linear mixed models (GLMMs) and inference is made using Markov chain Monte Carlo (MCMC) techniques. This general framework allows quite complex models where the random effects essentially play the role of unobserved latent factors influencing default and migration rates; to capture economic cycle effects the latent factors are allowed to have a dynamic time-series structure. An empirical study of Standard and Poors data shows strong evidence for economic cycles and also reveals pronounced sectoral heterogeneity in default and migration rates.