A continuous-time model of departure time choice for urban shopping trips

Abstract

ABSTRACT:This paper proposes a continuous-time hazard duration model for urban shopping trip departure time choice. The time frame for the analysis of departure time is the entire day. The continuous-time model uses a non-parametric baseline hazard distribution, employs a non-parametric representation for the time-varying effect of covariates, and accommodates time-varying covariates.

Flexible substitution patterns in models of mode and time of day choice: new evidence from the UK and the Netherlands

“…A related set of studies have modelled departure time jointly with route choice using model systems in which a continuously variable choice of departure time is linked to a discrete choice of route (e.g., Mannering et al, 1990;Mahmassani and Chang;, Mahmassani et al, 1991. More recently, interest has also developed in formulating continuous models of departure time choice within the framework of hazard based duration models (e.g., Bhat and Steed, 2002;Wang, 1996). In addition, models of individual trip departure time are also increasingly being embedded in more general models of activity choice and scheduling behaviour (see, e.g., Ashiru et al, 2003 and the work reviewed therein).…”

Prediction of Commuter’s Daily Time Allocation

“…Small [9] compared MNL, OGEV and Nested logit model in time of day modelling. As for duration prediction, Bhat and Steed [10] estimated a Hazard model of urban shopping trip durations. Juan and Xianyu [11] predicted daily travel time using the Hazard model.…”

On the Impacts of Telecommuting over Daily Activity/Travel Behavior: A Comprehensive Investigation through Different Telecommuting Patterns

“…Studies of trip departure times have been of interest to researchers as they provide an understanding of temporal distribution of daily trips in a 24-hour span (Abkowitz, 1981;Small, 1982;McCafferty and Hall, 1982;Hendrickson and Plank, 1984;Bhat, 1998aBhat, , 1998bSteed and Bhat, 2000;Bhat and Steed, 2002;Ettema and Timmermans, 2003;Jou, 2001;Jou et al, 2008;and Komma, 2008). Such studies are important for planning traffic control strategies, real-time operational information, and effectiveness of transportation demand management measures.…”

“…While early studies usually focused on discrete time-ofday intervals, there has been a shift towards treating departure time as a continuous variable. Bhat and Steed discussed a number of disadvantages of discrete time-of-day modeling (Steed and Bhat, 2000;Bhat and Steed, 2002), including the unstable model results due to ad-hoc temporal partitioning of the day, inconsistencies of the results at interval boundaries, and impediments imposed on further applications of the model in real world time-dependent strategies. As a substitute, survival models based on hazard functions are introduced and applied.…”

“…Several attempts have been done in order to adopt a parametric definition for the baseline hazard function. However, Bhat and Steed (2002) argued that the parametric approach will generally lead to an inconsistent estimation of the hazard function when the assumed parametric form is incorrect and that with a non-parametric form, the resulting estimates are consistent and the loss of efficiency (resulting from disregarding information about the hazard's distribution) may not be substantial.…”