In this paper we present the methodological framework of a dynamic discrete-continuous choice model (DDCCM) of car ownership, usage and fuel type. The approach consists of embedding a discrete-continuous choice model into a dynamic programming (DP) framework. This work proposes the following novel features. First, decisions are modeled at a household level. Second, we consider an extensive choice variable which involves the car replacement decision, the annual driving distance, the fuel type, the decision to take a company car, or a new versus second-hand car.