CEPAR

You are here

CEPAR Longevity Risk Workshop 2019

Data graphs

Image: Data chart in front of a skyline

27 November 2019
Mathews Building (Map Ref F23), Room 214, UNSW Sydney

This workshop, hosted by CEPAR, covered the latest developments in Longevity risk for early career researchers and practitioners. It also covered fraility models and mortality heterogeneity, data analytics techniques applied to mortality, continuous time mortality models designed for practical application, developments in variable annuities and computational techniques as well as hedging longevity risk with value-based longevity indices.

The Workshop was held on Wednesday 27 November, in conjunction with the 27th Colloquium on Pensions and Retirement Research that was held on 2-3 December 2019. 

 
Presenters
  • Professor Ermanno Pitacco - Professor of Actuarial Mathematics and Life Insurance Techniques, University of Trieste, Italy
  • Dr Andrés Villegas - Senior Lecturer in the School of Risk and Actuarial Studies, and Associate Investigator at CEPAR, UNSW, Sydney
  • Professor Michael Sherris - Professor of Actuarial Studies in the School of Risk and Actuarial Studies, UNSW, Sydney and Chief Investigator at CEPAR
  • Dr Runhuan Feng - Associate Professor of Mathematics, Director of Actuarial Science, University of Illinois at Urbana-Champaign, USA
  • Dr Jonathan Ziveyi - Senior Lecturer in the School of Risk and Actuarial Studies,  and Associate Investigator at CEPAR, UNSW, Sydney

Presenter's bioographies can be found here.

Workshop Program

8.45am

Arrival and Registration

9.00am

Ermanno Pitacco, University of Trieste

“From Gompertz to frailty models: mortality modeling for actuarial applications”.

This presentation will provide a survey on “mortality laws”, with special focus on actuarial applications. The survey will start from the seminal contribution by B. Gompertz, the first mortality model based on biological assumptions. Various generalizations of the Gompertz law will then be addressed. Special attention will be placed on heterogeneity in mortality, caused by both observable and unobservable risk factors. In particular, frailty models will be presented, and relevant implementations in the field of life annuities and disability insurance will be discussed.

10.00AM

Morning Tea

10.15AM

Andrés Villegas, UNSW Sydney

“Developments in the application of data analytics techniques to mortality modelling”

This presentation will discuss application of data analytics techniques to the modelling of mortality. We will start by providing a survey of the recent literature applying modern statistical learning techniques to the projection and forecasting of mortality rates and life expectancy. We will then focus on a case study looking at the use of regularisation and cross-validation techniques in the automatic construction of mortality projection models. We will finish by outlining possible future directions on the use of statistical learning and machine learning for improving mortality forecasts.

11.15AM

BREAK

11.30AM

Michael Sherris, UNSW Sydney

“Developments in multi-factor continuous time mortality modelling.”

This presentation will introduce continuous-time multi-factor mortality models covering the dynamics of mortality rates, closed-form expressions for survival curves. Application of models to age-period and age-cohort data as well as multi-cohort models will be introduced. The Kalman filter and estimation of the models will be covered highlighting how Poisson variation can be incorporated into the model estimation. A comparison of fits and prediction using historical US mortality data will be used to assess model performance. Applications of the models are discussed including quantifying price of mortality risk using Blackrock CORI indices.

12.30PM

LUNCH BREAK

1.30PM

Runhuan Feng, University of Illinois

“Fundamentals of variable annuity market, product designs and latest technical developments.”

Variable annuities are hybrids of traditional life insurance and financial derivatives. The challenges from the modeling and risk management of these products stem from complex guaranteed benefits, dynamic policyholder behavior, and the interaction of mortality and financial risks. Today's computational power and technology make it possible for the life industry to develop highly sophisticated models. Nonetheless, the demand for computational power continues to grow. While the industry continues to rely heavily on hardware innovations, we are approaching a crossroads about how to proceed. In this talk, we review fundamental principles underlying industrial modeling practice and latest development in the academia for modeling of new product features and computational efficiency.

2.30PM

Afternoon Tea

3.00PM

Jonathan Ziveyi, UNSW Sydney

“A value -based longevity index: a consistent framework for valuation of longevity-linked products and basis risk quantification"

In this presentation we propose a universal value-based longevity index constructed from blending a multi-population affine term structure model for mortality evolution along with a dynamic Nelson-Siegel model for the dynamics of interest rates. We examine effectiveness of the index in hedging retirement income portfolios by presenting illustrative examples with the aid of US economic and population data where we note significant reduction in basis risk relative to indices based purely on mortality rates. We also note that interest rate and inflation risks can also materially influence the value of longevity-linked liabilities.

4.00PM

WORKSHOP CONCLUDES


Direct event enquiries to: Amy Brushwood

Direct media enquiries to: Silke Weiss

Date: 
Wednesday, November 27, 2019 - 08:30
End date: 
Wednesday, November 27, 2019 - 16:30
Location: 
Room 214, Mathews Building (Map Ref F23), UNSW Sydney