Mortality surface by means of continuous time cohort models

We study and calibrate a cohort-based model which captures the characteristics of a mortality surface with a parsimonious, continuous-time factor approach. The model allows for imperfect correlation of the mortality intensity across generations. It is implemented on UK data for the period 1900–2008....

Full description

Saved in:
Bibliographic Details
Published in:Insurance, mathematics & economics mathematics & economics, 2013-07, Vol.53 (1), p.122-133
Main Authors: Jevtić, Petar, Luciano, Elisa, Vigna, Elena
Format: Article
Language:English
Subjects:
Age effect
Algorithms
Calibration
Cohort analysis
Cohort effect
Correlation
Correlation analysis
Demography
Differential analysis
Differential Evolution algorithm
Forecasts
Mortality
Mortality forecasting
Optimization algorithms
Stochastic models
Stochastic mortality
Studies
United Kingdom
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We study and calibrate a cohort-based model which captures the characteristics of a mortality surface with a parsimonious, continuous-time factor approach. The model allows for imperfect correlation of the mortality intensity across generations. It is implemented on UK data for the period 1900–2008. Calibration by means of stochastic search and the Differential Evolution optimization algorithm proves to yield robust and stable parameters. We provide in-sample and out-of-sample, deterministic as well as stochastic forecasts. Calibration confirms that correlation across generations is smaller than one. •We model the mortality surface with continuous-time, cohort-based stochastic intensities.•We specify to two-factor, Ornstein–Uhlenbeck intensities.•We model and calibrate correlation across different cohorts.•Fit of historical data is good, both deterministic and stochastic forecast reliable.•Differential-Evolution algorithm provides cohort correlations high but not perfect.
ISSN:0167-6687
1873-5959
DOI:10.1016/j.insmatheco.2013.04.005