March event: Careers in Statistics

All members, especially young statisticians, are welcome to the following event on careers in statistics taking place on Wednesday 26th March 2014, starting from 1pm.

This *free* event is designed to broaden the horizons of career-young statisticians through showcasing the wide range of different statistical careers that are available. Statistics covers fields as diverse as medicine, finance, industry, environment, government and academia. There will be a range of talks from these areas with invaluable networking opportunities.

Speakers are confirmed from Phastar, AstraZeneca, Sparx, NowGen, Department of Work and Pensions, and other companies based in northern England.

The event is intended for anyone about to embark on, or at the beginning of, their statistical career, so whether you have already begun to specialise in a particular area of statistics or are still studying, the day is sure to be of great value.

The event is *free of charge* but we do ask participants to register at http://www.eventbrite.co.uk/e/careers-in-statistics-registration-10313603265

Venue: Alan Turing Building, University of Manchester.
This is building is #46 on the campus map where you can also find travel information.:
http://www.manchester.ac.uk/aboutus/travel/maps/

Timetable:

1:00 – 1:30  Arrival

1:30 – 3:00  Session 1 – including talks from Royal statistical society, an environmental statistician, department of work and pensions, and a Professor of Medical Statistics.

3:00 – 3:30 Coffee and Refreshments

3:30 – 5:00 Session 2 – including talks from Astrazeneca, Phastar, NowGen and a local clinical trials unit

5:00 Wine reception and networking

This is jointly organised by University of Manchester School of Mathematics, and the Royal Statistical Society (Young Statisticians’ Section and Manchester Local Group).

Please direct any queries about the event to Matthew Sperrin:
matthew.sperrin@manchester.ac.uk

We look forward to seeing you in Manchester!

Organising Committee:
Peter Foster
Christiana Charalambous
Mark Pilling
Elizabeth Boggis
Matthew Sperrin

December seminar: Assessing Disclosure Risk in Sample Microdata

Date: Wednesday 11th December 2013

Time: 16:00-17:00 (tea/coffee and mince pies available from 15.30)

Venue: Room G207, Alan Turing Building, The University of Manchester, Manchester

Building 46 on the campus map: http://www.manchester.ac.uk/aboutus/travel/maps/interactive-map/

Assessing Disclosure Risk in Sample Microdata

Professor Natalie Shlomo
Professor of Social Statistics, The Cathie Marsh Centre for Census and Survey Research, University of Manchester

Abstract:
Disclosure risk occurs when there is a high probability that an intruder can identify an individual in released sample microdata and confidential information may be revealed.  For most social surveys, the population from which the sample is drawn is generally not known or only partially known through marginal distributions. The identification is made possible through the use of a key, which is a combination of indirectly identifying variables, such as age, sex and place of residence. Disclosure risk measures are based on the notion of population uniqueness in the key. In order to quantify the disclosure risk, probabilistic models are defined based on distributional assumptions about the population counts inferred from the observed sample counts. The parameters for the distribution are estimated through log-linear models.  The probabilistic framework is expanded to cover the case of misclassification on the key variables, either arising from the  survey process or as a  result of perturbative disclosure control techniques that may have been applied to the data. The methods are demonstrated on real data drawn from extracts of the 2001 United Kingdom Census.

This is joint work with Prof. Chris Skinner of the London School of Economics and Political Science.

Statistical Methods in Genetic Epidemiology

The Centre for Biostatistics at The University of Manchester and the Royal Statistical Society Manchester Local Group are organising the following seminar, which all are welcome to attend.

Further details are at: http://www.statslife.org.uk/events/events-calendar/icalrepeat.detail/2013/10/16/53/-/statistical-methods-in-genetic-epidemiology

Date: Wednesday 16th October 2013

Time: 14.00-17.00

Venue: Manchester Dental Education Centre (MANDEC), Higher Cambridge Street, Manchester, M15 6FH

Theme: Statistical Methods in Genetic Epidemiology

Programme:

14:00-14:50: Dr Frank Dudbridge, London School of Hygiene and Tropical Medicine

“Power and predictive accuracy of polygenic risk scores”
Polygenic scores have recently been used to summarise genetic effects among an ensemble of markers that are not individually significant.  Association between a trait and this composite score implies that a genetic signal is present among the selected markers, and the score can then be used for prediction of individual trait values.  Here I derive the statistical properties of the polygenic score from a quantitative genetics model in terms of the sizes of the two samples, explained genetic variance, selection thresholds for including a marker in the score, and method for weighting effect sizes in the score.  A novel approach to estimating the variance explained by a marker panel is also proposed.  I show that published studies with significant association of polygenic scores have been well powered, whereas those with negative results can be explained by low sample size.  I also show that useful levels of prediction may only be approached when predictors are estimated from very large samples, up to an order of magnitude greater than currently available.

14:50-15:40: Professor John Thompson, University of Leicester

“Beyond Mendelian Randomization”
Mendelian randomization (MR) is a form of instrumental variable analysis that seeks to establish causal associations between exposures and outcomes using non-randomized data. In such an analysis, genes are used as the instruments and so the recent growth in data from genomewide association studies the technique has made MR very popular. However, MR requires the assumption that the chosen genes do not have a pleiotropic effect on the outcome and this assumption is rarely justified. We will argue that, because of the likelihood of pleiotropy, a MR does not provide convincing evidence of causality between exposure and outcome, but that it does still provide useful information about common causal pathways shared by the exposure and the outcome. We will suggest that genes have an important role in the search for causation within epidemiological data, but that MR is a very special case of this general analysis that is currently being over-used.

15:40-16:00: Refreshments

16:00-16:50: Dr Vincent Plagnol, University College London

“Bayesian test for co-localisation between pairs of genetic association studies using summary statistics”
Genetic association studies, in particular the genome-wide association study (GWAS) design, have provided a wealth of novel insights into the aetiology of a wide range of human diseases. The next challenge consists of understanding the molecular basis of these associations. We have developed a novel statistical methodology to assess whether two association signals are consistent with a shared causal variant. An application is the integration of disease scans with expression quantitative trait locus (eQTL) studies, but any pair of GWAS datasets can be integrated in this framework. A key feature of the method is the ability to derive the output statistics from single SNP summary statistics, hence making it possible to perform systematic meta-analysis type comparisons across multiple GWAS datasets. We demonstrate the value of this approach by applying the methodology to several cardiovascular and autoimmune association studies.

Registration: For refreshment purposes, to register for this free event please contact Wendy Lamb on wendy.j.lamb@manchester.ac.uk  or +44 (0)161 275 5764

Wednesday 5th June 2013

Professor Martin Bland, University of York

Room 5.210, University Place, The University of Manchester

16:00-17:00 (tea/coffee and biscuits available from 15.30)

Improving statistical quality in published research: the clinical experience

Over the past 40 years, the quality of clinical research has improved greatly.  I shall try to show how this has come about and identify key factors in this improvement.  I shall go on to show that there is more to do.  I shall look at the position in non-clinical biomedical research and see whether there are any lessons to be drawn from the clinical experience.

Please note the change of venue from the original posting; this seminar will now take place in Room 5.210, University Place.  This can be found at Building Number 37 on The University of Manchester Campus Map.