Analysis of Vehicle Theft: A Case Study in Malaysia using Functional Forms of Negative Binomial Regression Models

作者：Zulkifli Malina; Ismail Noriszura^*; Razali Ahmad Mahir

来源：Applied Mathematics and Information Sciences, 2013, 7(2): 389-395.

• crime index
• vehicle theft
• Poisson regression
• negative binomial regression
• functional form
• overdispersion

This paper proposes the application of the two well known models of negative binomial regression, namely the NB-1 and the NB-2, and the functional form of negative binomial regression, namely the NB-P, for the analysis of vehicle theft crime. The advantage of using the NB-P is that it parametrically nests both the NB-1 and the NB-2, and allows statistical tests of the NB-1 and the NB-2 models against a more general alternative. In this study, the functional forms of negative binomial regression model were fitted to vehicle theft crime data obtained and compiled from ten insurance companies in Malaysia. The results of this study indicate that the risks of vehicle theft are higher for vehicles above eight years, vehicles above 1800 c.c., local vehicles and vehicles located in the central area of Malaysia.

• 出版日期2013-6