CASdatasets: Insurance datasets [R package fitdistrplus version 1.2-1]

CASdatasets: Insurance datasets

A collection of datasets, originally for the book 'Computational Actuarial Science with R' edited by Arthur Charpentier (CAS with R). Now, the package contains a large variety of actuarial datasets.

Version:	1.2-1
Depends:	R (>= 3.6.0), xts, survival
Imports:	lattice
Suggests:	sp, sf; knitr, kableExtra, quarto, rmarkdown; AER, boot, broom, ChainLadder, caret, demography, dplyr, forecast, ggplot2, glmnet, locfit, MASS, mgcv, pscl, rainbow, ranger, RColorBrewer, tidyr, tidyverse, wesanderson
Published:	2026-04-11
Author:	Christophe Dutang [aut, cre], Arthur Charpentier [aut], Ewen Gallic [ctb], Julien Siharath [ctb].
Maintainer:	Christophe Dutang <christophe.dutang at ensimag.fr>
License:	GPL-2 \| GPL-3 [expanded from: GPL (≥ 2)]
BugReports:	https://github.com/dutangc/CASdatasets/issues
URL:	https://dutangc.github.io/CASdatasets/
NeedsCompilation:	yes
Citation:	To cite CASdatasets in publications use: C. Dutang and A. Charpentier (2026), CASdatasets: Insurance datasets, R package version 1.2-1, DOI 10.57745/P0KHAG.
Materials:	NEWS
In views:	ActuarialScience

Documentation:

Reference manual:	CASdatasets.pdf
Overall vignettes:	CASdatasets overview CASdatasets' usage in research
Non-life insurance vignettes:	Poisson regression of a French MTPL dataset Quasi-Poisson regression of an Australian MTPL dataset Zero-inflated Poisson/negative binomial analysis of a Norwegian MTPL dataset Zero-inflated negative binomial analysis of a French MTPL dataset Generalized additive model analysis of a Belgian MTPL dataset Claims reserving of a French MTPL triangle
Other vignettes:	Calibration Metrics for RF and GLM Using French Insurance Data Reserving exercice on reporting delays in AIDS Data Mortality Table Analysis: A Study of French Mortality Data

Downloads:

Package source:	CASdatasets_1.2-1.tar.gz
Package installation:	Please make sure that xts, zoo packages are installed. Use this R repository within R via install.packages("CASdatasets", repos = "http://dutangc.perso.math.cnrs.fr/RRepository/pub/", type="source") library(CASdatasets) ?CASdatasets We recall that packages listed in Suggests field are needed only to build vignettes.

Usage:

Contests, case studies, benchmarks or education usage:

Kaggle contest and a corresponding notebook with GLM, Neural Nets and XGBoost,
ADS: the Swiss Association of Actuaries' Initiative, see, e.g., Nesting Classical Actuarial Models into Neural Networks published by Schelldorfer, Wuthrich (2019) on SSRN,
DAV: the Data Science Challenges of the German Association of Actuaries,
scikitlearn: benchmarks and tutorial for GLMs.
Datasets used in some actuarial programs : Dauphine Paris, France, DUAS Strasbourg, France, ENSAE Paris, France, ISFA Lyon, France, Univ. Laval Quebéc, Canada, UQAM Montréal, Canada,

Usage vignette:

CASdatasets' usage in research

TOP4 usage of freMTPL:

Noll A, Salzmann R, Wuethrich MV (2020). “Case study: French motor third-party liability claims.” SSRN Electronic Journal. doi:10.2139/ssrn.3164764.
Pocuca N, others (2020). “Modeling frequency and severity with zero-inflated generalized cluster-weighted models.” Insurance: Mathematics and Economics, 94, 1–16. doi:10.1016/j.insmatheco.2020.06.004.
Miljkovic T, Fernandez M (2018). “On Two Mixture-Based Clustering Approaches Used in Modeling an Insurance Portfolio.” Risks, 6(2), 57. doi:10.3390/risks6020057.
Denuit M, Charpentier A, Trufin J (2021). “Autocalibration and Tweedie-dominance for insurance pricing with machine learning.” Insurance: Mathematics and Economics, 101, 485–497. doi:10.1016/j.insmatheco.2021.09.001.