Statistical Laboratory Seminars
Lent Term 2002
Centre for Mathematical Sciences
Wilberforce Road, Cambridge, CB3 0WB
Tel: (01223) 337958
Fax: (01223) 337956
Email: [log in to unmask]
Seminar organizer, Susan Pitts. (01223) 337960
Seminars will be held in Meeting Room 12
All interested are welcome
Seminar Schedule:
Friday, 18 January
Mike Robinson (Man Investment Products)
Applications of statistics in finance -- development of a trading system
I currently work in the research team of Man Investment Products a company
which specialises in the field of alternative investment products. We
manage funds based on systematic trading strategies applied to a wide
range of international financial markets. Part of the research our team
does is in the development of the methods on which the trading strategies
are based. A large part of this development requires the use of
statistical techniques.
In this talk I will give an overview of the issues that must be addressed
in the development of a trading strategy and illustrate the major points
through application to daily data on the FTSE Allshare index. I will cover
the following points: the concept of futures markets which lend themselves
nicely to systematic trading methods; the modelling of temporal dependence
in both the mean and variance (volatility) of a time series; choice of
trading rule; the effect of trading costs; risk assessment; and finally
the benefits of diversification by trading more than one market.
Friday, 15 February
Shaun Seaman (MRC Biostatistics Unit, Cambridge)
Equivalence of Prospective and Retrospective Models in the Bayesian
Analysis of Case-Control Studies
The natural likelihood to use for a case-control study is a
`retrospective' likelihood, i.e. a likelihood based on the probability of
exposure given disease status. Prentice & Pyke (1979) showed that the
maximum likelihood estimates obtained from the retrospective likelihood
are the same as those obtained from the `prospective' likelihood
(i.e. that based on probability of disease given exposure). Until now, no
such result was available for the Bayesian analysis.
I shall show how Prentice & Pyke's result may be proved more easily and
how an analogous result may be obtained for the Bayesian analysis. That
is, I shall show that Bayesian analysis of case-control studies may be
done using a relatively simple model, the logistic regression model, which
treats data as though generated prospectively and which does not involve
nuisance parameters for the exposure distribution. This model can be
fitted using the software WinBUGS.
Friday, 22 February
Alan Welsh (University of Southampton)
Incomplete detection in enumeration surveys
Abstract to follow
|