ATTENTION/WARNING - NE PAS DÉPOSER ICI/DO NOT SUBMIT HERE

Ceci est la version de TEST de DIAL.mem. Veuillez ne pas soumettre votre mémoire sur ce site mais bien à l'URL suivante: 'https://thesis.dial.uclouvain.be'.
This is the TEST version of DIAL.mem. Please use the following URL to submit your master thesis: 'https://thesis.dial.uclouvain.be'.
 

Dimension-Reduction with Feed-Forward Neural Network Applied to Mortality

Haddi, Naïm
(2021)

Files

Haddi_80701400_2021.pdf
  • Open access
  • Adobe PDF
  • 1.92 MB
Download

Details

Supervisors
Hainaut, Donatien
Faculty
Faculté des sciences
Degree label
Master [120] en sciences actuarielles, à finalité spécialisée
Abstract
In the present work, a feed-forward neural network is implemented for a dimensionalty reduction purpose applied to mortality. Based on Professor D. Hainaut paper, this model is built suggesting a deviance instead of a mean square error as a loss function. In particular, it is a Poisson deviance that is implemented as the number of death follows a Poisson law. The Poisson model has the structure of log-mortality rates as the Lee Carter model. The dimension-reduction is performed on log-mortality rates with a non linear principal component analysis to model kappa(t) located in the bottleneck of the neural nets with kappa(t) representing the aging component of the Lee Carter model.