| Summary: | 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.
|
|---|