In this research we specify a dynamic discrete-continuous choice model (DDCCM) of car ownership and usage. This model embeds a discrete-continuous choice model into a dynamic programming (DP) framework. The DDCCM allows to jointly model the transaction type, the annual driving distance, the fuel type, the car ownership status and the car state (i.e. the purchase of a new or second-hand car) relative to each car in a households fleet. Moreover we model these decisions at the household level and for up to two cars per household. In this paper, the methodology is illustrated by an example of application where parameters are postulated.