6533b872fe1ef96bd12d39b6
RESEARCH PRODUCT
Modelling and forecasting mortality in Spain
Francisco MontesAna DebónFrancisco Puigsubject
Information Systems and ManagementLee–CarterGeneral Computer ScienceESTADISTICA E INVESTIGACION OPERATIVAManagement Science and Operations ResearchLee carterIndustrial and Manufacturing EngineeringDynamic life tablesMortality dataModeling and SimulationLife insuranceEconomicsEconometricsStatistical analysisDynamismBootstrap confidence intervalParametric statisticsForecastingBootstrap confidence intervalsdescription
[EN] Experience shows that static life tables overestimate death probabilities. As a consequence of this overestimation the premiums for annuities, pensions and life insurance are not what they actually should be, with negative effects for insurance companies or policy-holders. The reason for this overestimation is that static life tables, through being computed for a specific period of time, cannot take into account the decreasing mortality trend over time. Dynamic life tables overcome this problem by incorporating the influence of the calendar when graduating mortality. Recent papers on the topic look for the development of new methods to deal with this dynamism. Most methods used in dynamic tables are parametric, apply traditional mortality laws and then analyse the evolution of estimated parameters with time series techniques. Our contribution consists in extending and applying Lee–Carter methods to Spanish mortality data, exploring residuals and future trends.
| year | journal | country | edition | language |
|---|---|---|---|---|
| 2008-09-01 |