Motor Insurance Fraud Management Research
|Keywords||Motor Vehicle Claims Fraud LOGIST Regression Model FraudManagement|
The insurance industry as a sunrise industry is entering a period of vigorous development. In particular, the acceleration of urbanization, the increasing popularity of motor vehicles, motor vehicle insurance during last30years has been developing rapidly. This also provides stability for economic and social development and people’s living as an important safeguard. At the same time, motor vehicle insurance fraud phenomena frequently hit the newspapers. It can be said, the history of the motor vehicle insurance industry is as long as the motor vehicle insurance fraud phenomena.This study is on the basis of both domestic and foreign motor vehicle insurance fraud in empirical and theoretical studies. Firstly, the study discovers the root causes of insurance fraud, by analysis of the manifestation and its consequences. Then, countermeasures to address fraud will be proposed; Secondly, the dimension from the insurer’s practical experience, on the basis of A company’s status of vehicle insurance fraud management and its data base to apply of LOGIST regression model and to use SAS tools upon A company’s last five years auto insurance claims fraud. Also the single factor and two-factor analysis will be used to identify the main risk factor in claims fraud management. Finally, through the deep analysis and study of A company’s case, the study will raise new ideas for insurance industry and company levels to strengthen fraud management, and to achieve the significant change in the fraud risk management.The Study aims to provide comprehensive demonstration of motor vehicle insurance fraud management in the corporate level. This article by research on insurance fraud management not only provides a new management idea, but also technical support for the restructuring and development of the insurance company. However, it is difficult to identify the variables due to the limited data base and the quality of the overall data, which may have some effect on the model application in the study.