MULTI-STATE NON-HOMOGENEOUS SEMI-MARKOV MODEL OF DAILY ACTIVITY TYPE, TIMING AND DURATION SEQUENCE
Résumé
Understanding travelers' daily travel-activity pattern formation is an important issue for activity-based travel demand analysis. The activity pattern formation concerns complex interrelations between household members and individual's socio-demographic characteristics, which is in general difficult to quantify its influence. To investigate the interrelationship between travel-activity pattern formation and related covariates, we apply a non-homogeneous multi-state semi-Markov model to analyze the timing and duration of daily travel and activity sequence. We consider the observed travel-activity chains as a result of a semi-Markov process with a finite state space, i.e. the sequence of travel-activity depends only on its adjoining states. To estimate the effect of covariates related to individual and household characteristics, a Cox regression model is adopted in a semi-Markov process. The statistical tests of significance suggest that the duration of activity depends not only on its beginning time-of-day but also on the duration of travel/activity in prior episode. Empirical estimates of state transition hazard based on Belgian Mobility survey suggest useful insight on the relationship between the duration of activity and related socio-demographic characteristics.
Domaines
Economies et finances
Fichier principal
Multi-state_non-homogeneous_semi-Markov_hazard_model.pdf (495.43 Ko)
Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)