Data Science for Non-Life Insurance

The course puts focus on four main topics:

• basics of statistical and machine learning with a focus on the analysis of frequency and severity insurance data
• non-life insurance pricing methods, with GLMs, regression trees, random forests, gradient boosting machines and neural nets: technicalities, interpretation tools and surrogate models, model insights and resulting pricing models
• credibility theory and bonus-malus scales
• non-life claims reserving.

Link to the official course description.

Implementation in R and Python of pricing and reserving models will be covered.

Students will work on a group project: construction of a pricing model for a given data set.