# GEMAct: a Python package for non-life (re)insurance modeling

**Authors:** Gabriele Pittarello, Edoardo Luini, Manfred Marvin Marchione  
**arXiv:** [2303.01129](https://arxiv.org/abs/2303.01129)  
**Date:** 2023-03-02  
**Summary page:** https://www.summarizepaper.com/en/paper/2303.01129v2/

## Key points

- "GEMAct" is a Python package for actuarial modeling based on the collective risk model
- The library supports various applications such as costing and risk transfers, risks aggregation, and claims reserving
- New probability distributions have been added to those available in scipy, including the (a,b,0) and (a,b,1) discrete distributions as well as Archimedean and Elliptical classes of copulas
- GEMAct has a software implementation of the AEP algorithm which allows users to calculate cumulative distribution functions of dependent non-negative random variables given their dependency structure specified with a copula
- The theoretical framework underlying these actuarial models is introduced at the beginning of each section to provide readers with sufficient understanding.
- Overall, this paper presents an important contribution to actuarial modeling by providing a comprehensive library that supports various applications in non-life insurance modeling.

## Summary

The paper "GEMAct: a Python package for non-life (re)insurance modeling" introduces the gemact library, which is a Python package designed for actuarial modeling based on the collective risk model. The library provides support for various applications such as costing and risk transfers, risks aggregation, and claims reserving. Additionally, the authors have added new probability distributions to those available in scipy, including the (a,b,0) and (a,b,1) discrete distributions as well as Archimedean and Elliptical classes of copulas. One of the key features of gemact is its software implementation of the AEP algorithm. This algorithm allows users to calculate cumulative distribution functions of dependent non-negative random variables given their dependency structure specified with a copula. The theoretical framework underlying these actuarial models is introduced at the beginning of each section to provide readers with sufficient understanding. Overall, this paper presents an important contribution to actuarial modeling by providing a comprehensive library that supports various applications in non-life insurance modeling. The addition of new probability distributions and copulas expands upon existing tools available in scipy and enhances the capabilities of gemact. Moreover, its software implementation of AEP algorithm adds value to this package by allowing users to calculate complex probabilities efficiently.