01514292

Component

Modeling the activity scheduling problem as a chain of trips in a dynamic discrete choice framework gives a dynamically consistent and microeconomically sound model. However, the possibility to estimate such a model has previously been limited by the large number of states that are needed to represent all possible locations, times and activity combinations for an individual during a day. To overcome this problem the authors utilize that the state-space shrinks in the end of every day, and introduce some assumptions under which a sequential estimation approach can be used to get consistent estimates. The estimation approach involves: firstly, using sampling of alternative sequences of actions within a day to estimate part of the structural parameters; and secondly, conditioned on these parameters solve the dynamic discrete choice model for the between days choices using a nested fixed point algorithm. To verify the approach, the authors construct a simple prototype problem from which synthetic data can be simulated. Statistical tests verify that the sequential estimation approach gives consistent and unbiased estimates to all structural parameters, including the between day discount factor.

This paper was sponsored by TRB committee ADB40 Transportation Demand Forecasting. Alternate title: Within- and Between-Day Activity Scheduling by Infinite Trip Chaining.

01503729

14-0725

English

16p

2014

Transportation Research Board 93rd Annual Meeting

Digital/other

Figures; References; Tables

Microeconomics; Sequential analysis; Trip chaining

Activity scheduling; Discrete choice models; Nested algorithms

Highways; Planning and Forecasting; I72: Traffic and Transport Planning

Transportation Research Board Annual Meeting 2014 Paper #14-0725

TRIS, TRB, ATRI

Jan 27 2014 2:19PM