Modelling car ownership in a developing country: empirical data from Akure, Nigeria

Car ownership was examined in Akure, Nigeria. Car ownership was predicted using discriminant analysis as a probabilistic modelling technique. The factors that influence household car ownership were studied. Poisson distribution was used to determine the probability and the number of families that will own a certain number of cars in the study area. The discriminant and Poisson statistics predictions were compared with empirical ownership patterns for no car, one car, two car and three car ownership. The models predicted a reduction in the number of families without cars and an increase in the number of car-owning families. The importation of used cars into the country, employment, education and marital status influenced the car ownership.