Methods and statistical models used to identify uninsured car owners

Wojciech Bijak, Piotr Dziel, Stanisław Garstka and Krzysztof Hrycko
Additional contact information
Wojciech Bijak: Szkoła Główna Handlowa w Warszawie, Ubezpieczeniowy Fundusz Gwarancyjny
Piotr Dziel: Ubezpieczeniowy Fundusz Gwarancyjny
Stanisław Garstka: Ubezpieczeniowy Fundusz Gwarancyjny
Krzysztof Hrycko: Ubezpieczeniowy Fundusz Gwarancyjny

Collegium of Economic Analysis Annals, 2015, issue 37, 199-228

Abstract: The paper presents the approach used by the Polish Insurance Guarantee Fund to tackle the phenomenon of uninsured car owners. The process covers different areas concerning: 1. algorithms which identify discontinuity in MTPL coverage, 2. the usage of statistical modelling called supervised learning, including such models as: generalised linear models, decision trees and neural networks, 3. the cooperation with insurers to identify the uninsured. The paper presents the results obtained in exemplary statistical models, as well as the achieved accuracy of prediction. The publication presents further evolution of the developed system.

Keywords: MTPL insurance coverage control; automatic identification of the uninsured; supervised learning (search for similar items in EconPapers)
Date: 2015
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
http://rocznikikae.sgh.waw.pl/p/roczniki_kae_z37_09.pdf Full text (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:sgh:annals:i:37:y:2015:p:199-228

Access Statistics for this article

Collegium of Economic Analysis Annals is currently edited by Joanna Plebaniak, Beata Czarnacka-Chrobot

More articles in Collegium of Economic Analysis Annals from Warsaw School of Economics, Collegium of Economic Analysis Contact information at EDIRC.
Bibliographic data for series maintained by Michał Bernardelli ().