Actuarial datasets and projects for machine learning

Hi,

I am a CAS university liaison at a smaller state university that only has a few students possibly interested in an actuarial career, and we don’t offer an Actuarial Science major or minor.

The math chairperson is exploring the idea of a data science minor, and I think he would be open to some actuarial-type projects in some of the upper-level probability or machine learning courses. Are there some elementary actuarial data analysis datasets and projects that you can share with me? I am not looking for something of the level of the Roosevelt Mosley monograph or the CAS Hackathon. I don’t have access to what in the CAS Syllabus that could be useful. But if students are learning logistic regression and random forests, they can just as well learn these with predicting insurance claims. Prefer property-casualty.

Any suggestions? Thanks.

I know this exists, but haven’t done a ton with it myself. Some actuarial modeling courses utilize this R data though:

There is a link to the documentation of the data itself on this site.

Great, thank you. It looks like the good stuff is here GitHub - MHaringa/insurancerating: R-package for actuarial pricing
and
https://cran.r-project.org/web/packages/insurancerating/insurancerating.pdf

Pru is doing a bunch of ai things. Googliing prudential ai brings up a lot of hits.

1 Like