Vehicle Insurance Claim Data Study and Forecasting Model using Artificial Neural Networks

Author: 

Tornike Mzhavia

Degree: 

M.Sc.

Supervisors: 

Date: 

Thursday, June 2, 2016

Thesis language: 

English

Abstract: 

This thesis represents the work done for the project, with the main aim of studying real life vehicle insurance data for finding correlations and causations which lead to the insurance claims made by the customers. The aim of this project was to, if possible, create a predictive model based on Artificial Neural Network technology which would be trained based on the said database in order to successfully forecasting the risks of claims and creating fair ratemaking for the insurance company and their customers.

Electronic version: