RISK FACTORS SELECTION WITH DATA MINING METHODS FOR INSURANCE PREMIUM RATEMAKING

Insurance companies that have adopted the application of data mining methods in their business have become more competitive in the insurance market. Data mining methods provides the insurance industry with numerous advantages: shorter data processing times, more sophisticated methods for more acc...

Full description

Bibliographic Details
Main Authors: Amela Omerašević, Jasmina Selimović
Format: Article
Language:deu
Published: Faculty of Economics University of Rijeka 2020-12-01
Series:Zbornik radova Ekonomskog fakulteta u Rijeci : časopis za ekonomsku teoriju i praksu
Subjects:
glm
Online Access:https://www.efri.uniri.hr/upload/Zbornik%202_2020/12-Omerasevic_et_al-2020-2.pdf
Description
Summary:Insurance companies that have adopted the application of data mining methods in their business have become more competitive in the insurance market. Data mining methods provides the insurance industry with numerous advantages: shorter data processing times, more sophisticated methods for more accurate data analysis, better decision-making, etc. Insurance companies use data mining methods for various purposes, from marketing campaigns to fraud prevention. The process of insurance premium pricing was one of the first applications of data mining methods in insurance industry. The application of the data mining method in this paper aims to improve the results in the process of non-life insurance premium ratemaking. The improvement is reflected in the choice of predictors or risk factors that have an impact on insurance premium rates. The following data mining methods for the selection of prediction variables were investigated: Forward Stepwise, Decision trees and Neural networks. Generalized linear models (GLM) were used for premium ratemaking, as the main statistical model for nonlife insurance premium pricing today in most developed insurance markets in the world
ISSN:1331-8004
1846-7520