Presentation of EAJ Issue 13/1 - May 22nd: Optimal portfolios with sustainable assets: aspects for life insurers - presented by Prof. Dr. Ralf Korn | Micro-level prediction of outstanding claim counts based on novel mixture models and neural networks - presented by Alexander Rosenstock | An incremental loss ratio method using prior information on calendar year effects - presented by Dr. Ulrich Riegel | Generalized PELVE and applications to risk measures - presented by Dr. Anna Maria Fiori | On some effects of dependencies on an insurer’s risk exposure, probability of ruin, and optimal premium loading - presented by Ragnar Gudmundarson | Long-Term Stability of a Life Insurer's Balance Sheet - presented by Dr. Maximilian Diehl | Smooth projection of mortality improvement rates: A Bayesian two-dimensional spline approach - presented by Dr Mike Zhu | Neural networks meet least squares Monte Carlo at internal model data - presented by Dr. Christian Jonen | A public micro pension programme in Brazil: heterogeneity among states and setting up of a benefit age adjustment  - presented by Prof. Dr. Renata Gomes Alcoforado | Does autocalibration improve goodness of lift? - presented by Harrison Verelst | Model selection with Gini indices under auto-calibration - presented by Prof.  Mario V. Wüthrich

Presentation of EAJ Issue 13/1 - May 22nd

Alexander RosenstockDr. Anna Maria FioriDr. Christian JonenHarrison VerelstProf.  Mario V. WüthrichDr. Maximilian Diehl+5
Slide at 51:38
european actuarial
The proof is in the pu
academy
-0.5
Sobol Pf 1 CR Pf 1
Sobol Pf 2 CR Pf 2
Sobol Pf 3
CR Pf 3
Share slide
Summary (AI generated)

In our results, it is evident that our Neural Networks ensemble outperforms the polynomial approach. The outcomes we obtain are extremely strong.