You are here

Monday 2 December 2019 12:30 - 18:00

Purpose
The workshop will give participants detailed insights into how a large and detailed dataset can be used to assess levels of mortality inequality in England, using a number of novel statistical methods. We will also address the question: are regional variations in mortality real or imaginary?

Technical Content
The workshop will focus on a large dataset obtained from the UK’s Office for National Statistics (ONS) and related sources. Data are available at the level of LSOA’s – small geographical areas with, typically, 1000-2000 residents and include death counts, exposures and a significant number of socio-economic variables including the index of multiple deprivation (IMD). The workshop will look at how the socio-economic and other variables can be used to predict high or low mortality relative to national mortality.

The programme will include: an overview of the ONS data; stochastic modelling using IMD deciles; the application of Generalised Linear Models (GLM’s) to model socio-economic differences in mortality; non-linear predictive modelling techniques for socio-economic and geographical variables; using cause-of-death information as an additional outcome to better understand trends and inequalities.

Level of detail
The workshop aims to give participants a good understanding of the underlying statistical methods and their use in the assessment of mortality inequalities using a variety of socio-economic predictive variables.

Target audience
Life, pensions and specialist mortality actuaries; data analysts; actuarial students.

Prior knowledge
A good knowledge of statistics and some knowledge of survival analysis would be beneficial but is not essential.

Speakers
Professor Andrew John George Cairns FRSE, Professor of Financial Mathematics, Heriot-Watt University became the first Director of the Actuarial Research Centre in 2012, the international research arm of the Institute and Faculty of Actuaries. His research is centred around long term financial risk modelling and management for life insurers and pension plans with a particular focus on the financial risks associated with ever-increasing lifetimes.

Dr Torsten Kleinow, Associate Professor at Heriot Watt University, Edinburgh

Jie Wen is a PhD Student at Heriot Watt University, Edinburgh

 

12.30-12.45 Registration
12.45-13.00 Welcome and introduction
Andrew Cairns
13.00-13.20 What types of socio-economic data are publicly available for mortality data?      
Andrew Cairns
13.20-14.10 Multi-population stochastic mortality models: using the Index of Multiple Deprivation (IMD) deciles to model sub-population mortality
Torsten Kleinow
14.10-14.25 Refreshment break
14.25-15.15 Using Generalised Linear Models (GLM’s) to model the impact of socio-economic effects            
Jie Wen
15.15-16.05 Using local linear regression to model socio-economic and geographical effects
Andrew Cairns
16.05-16.20 Refreshment break
16.20-17.10 Mortality data by socio-economic group, region and cause of death: what do the patterns tell us?
Andrew Cairns
17.10-18.00 Discussion and refreshments

 

Location

Address

Staple Inn Hall, High Holborn, London, WC1V 7QJ

Nearest Public Transport

Chancery Lane Station