In the literature models simulating alternative choice probabilities are generally unable to explicitly simulate the variation in choice probability, due to a variety of events that affect the system characteristics of users and the transportation network. These aspects must be explicitly considered in order to simulate several choices such as path choice for high frequency service, evacuation and vehicle ownership. In these cases, dynamicmodels should be adopted. In this work a classification of dynamic modelsis proposed, defining dynamic models as regards attributes, random residuals or both aspects. Particular attention is devoted to sequential models,a kind of dynamic model resulting from sequential analysis. Sequential analysis of sequential recorded data can provided an additional level of information about whatever behaviour in comparison with non-sequential analysis. In this work, data related to the transition of a number of vehicles owned by a sample of households are analyzed according to the sequential approach. Sequential models to simulate vehicle ownership choices are proposed. For the first time, a probit distribution is proposed, allowing for any arbitrary variance-covariance structure of the disturbance term, as regards alternatives in present time and compared with the logit distribution.Results obtained by model experimentation are generally satisfactory. Forthe covering abstract see ITRD E145999
Abstract