3

I really need a dataset about automobile insurance claims to train and test learning algorithms.

I found references to Masachussets PIP claims data and to Spanish claims data in many scientific articles, but I couldn't find them...

The best would be to find claims which concern just insurance third party liability extensions: I mean theft, fire, acts of vandalism, atmospheric agents. But also claims concerning third party liability insurance would be great.

  • Did you find any dataset from this inquery? could you please share it? @Joe San Pietro is there any data description avaiable for this dataset (Auto Insurance Claims - Automobile Insurance claims including location, policy type and claim amount). I am struggling with the diff between 'claim amount' and 'Total Claim Amount' for instance. Thanks! – D.G May 13 at 8:58
6

A couple of new automobile insurance claim data sets have become available since this question was asked. Thought I'd list them here:

Emcien:

Published: Auto Insurance Claims - Automobile Insurance claims including location, policy type and claim amount http://dyzz9obi78pm5.cloudfront.net/app/image/id/560ec66d32131c9409f2ba54/n/Auto_Insurance_Claims_Sample.csv

Allstate:

Held a "claims severity" competition on Kaggle. https://www.kaggle.com/c/allstate-claims-severity

3

The book “Data: A Collection of Problems from Many Fields for the Student and Research Worker” by Andrews and Herzberg has such a data set as:

Table 68.1 Third Party Motor Insurance for Sweden, 1977 (112642 bytes)

When there is an answer to this question: The data from "Data: A Collection of Problems from Many Fields for the Student and Research Worker" by Andrews and Herzberg you can find the data from there.

And here is a direct link for the data: http://www.statsci.org/data/general/motorins.txt

1

You can find several datasets for R here, for the book Computational Actuarial Science with R.

Package ‘CASdatasets’

A completed project by the Insurance Risk and Finance Research Centre (www.IRFRC.com) hasassembled a unique dataset from Large Commercial Risk losses in Asia-Pacific (APAC) coveringthe period 2000-2013. The data was generously contributed by one global reinsurance companyand two large Lloyd’s syndicates in London. This dataset is the result of the project co-lead by DrMilidonis (IRFRC and University of Cyprus) and Enrico Biffis (Imperial College Business School),which can be referred to as the IRFRC LCR Dataset.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.