5

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.

StackUser
  • 231
  • 1
  • 2
  • 7
  • 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 '19 at 08:58

3 Answers3

7

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

Joe San Pietro
  • 71
  • 1
  • 1
4

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

Community
  • 1
kjetil b halvorsen
  • 409
  • 3
  • 14
2

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.

philshem
  • 17,647
  • 7
  • 68
  • 170
John Smith
  • 121
  • 1