LocalGLMnet: A Deep Learning Architecture for Actuaries

35 Pages Posted: 3 Sep 2021

Date Written: August 4, 2021

Abstract

The purpose of this tutorial is to discuss the LocalGLMnet architecture which is tailored to the needs of actuaries. The LocalGLMnet is a flexible network architecture for tabular data that allows for variable selection, the study of interactions, gives nice interpretations and allows to rank variable importance. We explore a LocalGLMnet on accident insurance claims data for which we also have short claim descriptions. In a second step we try to understand the predictive power of these claim descriptions by adding a recurrent neural network layer to process the claim texts into tabular data.

Keywords: LocalGLMnet, neural network, deep learning, variable selection, interactions, explainable artificial intelligence, XAI, generalized linear model, GLM, tabular data, variable importance, Shapley additive explanation, natural language processing, NLP, text recognition, recurrent neural network, LSTM

JEL Classification: G22, C45, C02, C12

Suggested Citation

Schelldorfer, Jürg and Wuthrich, Mario V., LocalGLMnet: A Deep Learning Architecture for Actuaries (August 4, 2021). Available at SSRN: https://ssrn.com/abstract=3900350 or http://dx.doi.org/10.2139/ssrn.3900350

Jürg Schelldorfer

Swiss Re

Mythenquai 50/60
Zurich, 8022
Switzerland

Mario V. Wuthrich (Contact Author)

RiskLab, ETH Zurich ( email )

Department of Mathematics
Ramistrasse 101
Zurich, 8092
Switzerland

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
262
Abstract Views
686
rank
147,061
PlumX Metrics