Detection of Automobile Insurance Fraud With Discrete Choice Models and Misclassified Claims

Abstract

The insurance industry is concerned with the detection of fraudulent behavior. The number of automobile claims involving some kind of suspicious circumstance is high and has become a subject of major interest for companies. This article demonstrates the performance of binary choice models for fraud detection and implements models for misclassification in the response variable. A database from the Spanish insurance market that contains honest and fraudulent claims is used. The estimation of the probability of omission provides an estimate of the percentage of fraudulent claims that are not detected by the logistic regression model.

Perceived Ethicality of Insurance Claim Fraud: Do Higher Deductibles Lead to Lower Ethical Standards?

“…Indeed, even though national legislation has focused mainly on insurer fraud (i.e., from within the insurance organization), many states have passed laws targeted at policyholder-based insurance fraud (Todd et al, 1999). However, the majority of the approaches to reducing insurance fraud have focused on identifying the perpetrators of such behavior (e.g., Artis et al, 2002;Schiller, 2006) rather than understanding why policyholders commit insurance fraud and thus, providing knowledge that allows fraud to be approached from an attitudeor behavioral-change perspective. This is despite the fact that public attitudes toward insurance fraud have been identified as one of the main obstacles to reducing this fraudulent activity and the need to change the public's perceptions is seen as a potential, albeit partial, solution (Carris and Colin, 1997;Smith, 2000).…”

FraudBuster: Reducing Fraud in an Auto Insurance Market

“…Belhadji et al 9 propose a procedure to identify and select the most significant features by measuring their sensitivity and specificity in detecting fraudulent claims. Artís et al 10 used a modified logit-based model to demonstrate the predictability and interpretability of binary discrete choice models for predicting auto insurance fraud. Caudill et al 11 extend this model in the presence of missing information.…”

Expiration Dates in Automobile Insurance Contracts: The Curious Case of Last Policy Month Claims in Taiwan

“…Cummins and Tennyson (1996) find that attitudes towards fraud significantly affect automobile liability claims. Other papers, such as those by Weisberg and Derrig (1998), Tennyson and Salsas-Forn (2002), Artis et al (2002) and Caudill et al (2005) develop techniques to identify or classify fraudulent claims. 5 If there is a plausible link between claim-related regulations or social environmental factors and claims or claim payments, we may suppose that opportunistic fraud or moral hazard is involved.…”