A Heteroscedastic Polarized Logit Model to Investigate the Competition of Bicycle for the Bikeable Trips with the Other Modes
- 20 September 2018
- journal article
- research article
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2672 (49) , 58-68
Abstract
The paper proposes a new discrete choice model, named the Heteroscedastic Polarized Logit (HPL) to investigate choice contexts with one or more alternatives with remarkably low market shares. The proposed model is used to investigate the factors influencing the choice of a bicycle as a travel mode in the National Capital Region (NCR) of Canada. Data from the latest household travel survey of the NCR are used to investigate the mode choices of bikeable trips. Bikeable trips are defined as trips with lengths shorter than 16 km as this is the observed maximum limit of a bicycle trip in the dataset. A large dataset with over 40,000 trip records is used for empirical investigation where the bicycle has the lowest mode share of 3%. The HPL model clearly shows its appropriateness and superiority over comparable models in such a context. The choice to walk is found to be more sensitive to trip length than the choice to cycle, yet walking is found to have three times larger market share than that of cycling. Similarly, motorized modes are found to have low sensitivity to travel time and other impedances and have larger market shares. Women and students are found not to prefer the bicycle as a travel mode. Cycling infrastructure is seen to be effective in increasing the choice of the bicycle as a travel mode, but it also becomes clear that additional soft policy initiatives would be necessary to increase the popularity of cycling among young people, students, and women.Keywords
This publication has 12 references indexed in Scilit:
- Myopic choice or rational decision making? An investigation into mode choice preference structures in competitive modal arrangements in a multimodal urban area, the City of TorontoCanadian Journal of Civil Engineering, 2016
- Two New Prediction-Driven Approaches to Discrete Choice PredictionACM Transactions on Management Information Systems, 2012
- Current issues in mode choice modelingTransportation, 2011
- Modelling commuting mode choice with explicit consideration of carpool in the choice set formationTransportation, 2011
- A Microeconomic Interpretation of the Maximum Entropy Estimator of Multinomial Logit Models and Its Equivalence to the Maximum Likelihood EstimatorEntropy, 2010
- Including Minor Modes of Transport in a Tour-Based Mode Choice Model with Household InteractionsJournal of Transportation Engineering, 2009
- Discrete Choice with Social InteractionsReview of Economic Studies, 2001
- Statistical Comparison of Non-Nested Probabilistic Discrete Choice ModelsTransportation Science, 1983
- Discrete choice theory, information theory and the multinomial logit and gravity modelsTransportation Research Part B: Methodological, 1983
- The dogit modelTransportation Research Part B: Methodological, 1979