Veranstaltungen

28. Mai 2024, 16:30 bis 18:00

Actuarial Modelling Club: „Cutting-edge in Pricing: Penalized Regression, Market Price Insights, and Practical Lessons“

Vortrag

Florian Nuding, Wüstenrot,

Bartosz Gaweda, Milliman,

Jan Küthe, Akur8

In most modern insurance companies, actuaries are debating between two main pricing methods: GBMs and GLMs. We will demonstrate how penalized regressions in a GLM framework can effectively tie standard Credibility practices and incorporate some of the benefits of GBMs, our talk aims to contribute to the fusion of these techniques as solid alternatives to standard GLMs for rate making.

Historically, this technique was popularized by the machine learning literature, and how it is taught is not synchronized with how actuaries approach modeling to solve insurance problems. This should not be the case: the penalized framework is versatile and effectively solves many insurance use cases that are currently tackled via established techniques.

First, we will see how, practically, and theoretically, Penalized Regressions are in essence Credibility procedures and allow to blend GLM with credibility to reduce overfitting and improve a model's ability to generalize.

On the other hand, we will display how Penalized regression can be thought of as Gradient Boosting Machines (GBMs), a powerful yet completely black box modeling technique.

Modern Modelling & Portfolio evaluation from a practical perspective, in the Austrian market.

From the point of view of a dynamic local insurer, we would like to discuss smart ways to identify segments with varying degrees of profitability and provide suggestions on the implementation of strategies in this setup. To do this fast and efficiently, we will present how we can leverage machine learning to speed up actuarial work.

Lastly, drawing from the successful projects undertaken by Milliman Italy, we will demonstrate the substantial value added by market benchmark prices to enhance insurance decision-making.

Logo of the research unit Financial and Actuarial Mathematics (FAM) at TU Wien

FAM @ TU Wien

Kalendereintrag

Veranstaltungsort

Freihaus der TU Wien
1040 Wien
Wiedner Hauptstraße 8-10

 

Veranstalter

FAM @ TU Wien gemeinsam mit Anerkannten Aktuaren der AVÖ
FAM-office
fam@fam.tuwien.ac.at

 

Info-Link

https://fam.tuwien.ac.at/vr/20240528.php

 

Öffentlich

Ja

 

Kostenpflichtig

Nein

 

Anmeldung erforderlich

Ja

Anmeldelink auf Veranstaltungsseite