In this paper we propose a stategy to deal with outliers and nonlinearities in mortality rates in order to improve the predictive capacity of the LC model. In particular we use the singular value decomposition based on density power divergence to account for outliers and auto-regressive neural networks to handle nonlinearities. Empirical evidence on real data shows a significant improvement contribution of the proposed methodology.
A Robust Non Linear Lee-Carter Type Model for Mortality Rate
La Rocca, Michele;Perna, Cira;Sibillo, Marilena
2026
Abstract
In this paper we propose a stategy to deal with outliers and nonlinearities in mortality rates in order to improve the predictive capacity of the LC model. In particular we use the singular value decomposition based on density power divergence to account for outliers and auto-regressive neural networks to handle nonlinearities. Empirical evidence on real data shows a significant improvement contribution of the proposed methodology.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
