Comparing Survival Analysis and Discrete Choice Specifications Simulating Dynamics of Vehicle Ownership

Modeling travel-related decisions of transport system users is the core of many behavioral travel demand models. It is of great significance to use these models for planning purposes in which decisions of individuals are simulated for specific time intervals or on a continuous dimension until the target year. Discrete choice and survival analysis methods are two popular econometric structures to model and forecast time-dependent outcomes. This paper elaborates the conceptual and practical differences between these two methods in the context of vehicle ownership modeling. There are meaningful differences between these two methods including data preparation approaches, interpretations of the variable of interest in the model, and the simulation procedures. Further, this paper shows how negligent application of discrete choice methods for modeling time-to-event variables results in specification bias. This discussion paves the path for using hazard-based models in travel demand modeling, as the application of these models have been quite limited compared with their capacity.

• Record URL:
  https://doi.org/10.1177/0361198118791911
• Availability:
  • Find a library where document is available. Order URL: http://worldcat.org/issn/03611981
• Authors:
  • Ghasri, Milad
  • Rashidi, Taha Hossein
  • Saberi, Meead
• Publication Date: 2018-12

Language

• English

Media Info

• Media Type: Print
• Features: Figures; References; Tables;
• Pagination: pp 34-45
• Serial:
  • Transportation Research Record: Journal of the Transportation Research Board
  • Volume: 2672
  • Issue Number: 49
  • Publisher: Sage Publications, Incorporated
  • ISSN: 0361-1981
  • EISSN: 2169-4052
  • Serial URL: http://journals.sagepub.com/home/trr

Subject/Index Terms

• TRT Terms: Automobile ownership; Choice models; Mathematical models; Simulation
• Subject Areas: Economics; Highways; Planning and Forecasting; Society;

Filing Info

• Accession Number: 01660377
• Record Type: Publication
• Report/Paper Numbers: 18-02489
• Files: TRIS, TRB, ATRI
• Created Date: Feb 20 2018 9:28AM