Over the last century the human mortality has declined globally. The changes in mortality trends strongly impact on pricing and reserve allocation of insurers and pension systems. Longevity risk derives from systematic deviation from the expected number of death; it has to be properly forecasted and managed. To this aim, researchers and practitioners make predictions resorting to classical demographic frameworks based on traditional extrapolative models. This chapter illustrates how ML can be used to improve fitting and forecasting of mortality. We present a numerical application based on real mortality data and use the forecasted mortality rates are to price two life insurance products and describe the impact of longevity on the actuarial reserves.
Improving longevity risk management through machine learning / Levantesi, S.; Nigri, A.; Piscopo, G.. - (2021), pp. 37-56. [10.4324/9781003037903].
Improving longevity risk management through machine learning
Levantesi, S.;Nigri, A.;
2021
Abstract
Over the last century the human mortality has declined globally. The changes in mortality trends strongly impact on pricing and reserve allocation of insurers and pension systems. Longevity risk derives from systematic deviation from the expected number of death; it has to be properly forecasted and managed. To this aim, researchers and practitioners make predictions resorting to classical demographic frameworks based on traditional extrapolative models. This chapter illustrates how ML can be used to improve fitting and forecasting of mortality. We present a numerical application based on real mortality data and use the forecasted mortality rates are to price two life insurance products and describe the impact of longevity on the actuarial reserves.File | Dimensione | Formato | |
---|---|---|---|
Levantesi_essentials-of-machine-learning_2021.pdf
solo gestori archivio
Tipologia:
Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza:
Tutti i diritti riservati (All rights reserved)
Dimensione
5.79 MB
Formato
Adobe PDF
|
5.79 MB | Adobe PDF | Contatta l'autore |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.