On behalf of Jan Küthe from Akur8 , the Board of the Polish Actuarial Association is pleased to invite its members, as well as members of the CEE actuarial associations, to a complimentary webinar in the „Open Floor” series, titled „On Similarities of Penalized Regression and GBM Modelling Approaches.”
About the webinar
Penalized regression is steadily becoming a mainstream application in ratemaking. There is momentum in the insurance space with innovation in research, software and production on penalized techniques that build and innovate the standard GLM models.
Historically, this technique was popularized by the machine learning literature, and how it is taught is not synchronized with how actuaries approach modelling to solve insurance problems. This should not be the case: the penalized framework is versatile and allows to effectively solve many insurance use cases that are currently tackled via established techniques.
First, we will see how, practically and theoretically, Penalized Regressions are effective 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 GBMs, a powerful but yet completely black box modelling technique.
Since Penalized regressions can effectively tie standard Credibility practices and incorporate some of the benefits of GBMs, this presentation aims to contribute to the diffusion of these techniques as solid alternatives to standard GLMs for ratemaking.
About the speaker:
Jan Küthe is an Actuary (DAV) from Germany and works at Akur8 as an Actuarial Data Scientist to help insurance companies unlock the potentials of the twenty-first century. Before that he has been working as an Actuarial Consultant for three years. He holds a Masters degree in Mathematics from the University of Bonn and is an avid reader of the books of Anna Seghers and Dietmar Dath.
When: 29 January 2025 (Wednesday) at 17.00 CET, Zoom meeting (https://zoom.us/j/97482466161 with no password or registration required).
Duration: the webinar is expected to last 60-70 minutes, including the Q&A.
Technical remarks:
