Travel mode choices for trips in the Montreal–Toronto corridor. Choice sets vary from two to four available modes among air, bus, car, and train. Monetary values were converted to Euro at an exchange rate of 1.6151 Canadian dollars = 1 Euro.
This data can be used to model the choice between the available air, bus, car, and train modes based on distance, cost, in-vehicle time, out-of-vehicle time, service frequency, household income, and the number of urban trip endpoints.
Format
A tibble with 15520 rows and 10 columns:
- case [
integer] The choice occasion identifier.
- mode [
character] The available travel mode.
- choice [
integer] Whether the mode was chosen.
- distance [
numeric] The trip distance in kilometers.
- cost [
numeric] The monetary cost of the mode in euros.
- inVehicleTime [
numeric] The time spent in the vehicle, in minutes.
- outVehicleTime [
numeric] The time spent outside the vehicle, in minutes.
- frequency [
numeric] The number of scheduled services; car has value 0.
- income [
numeric] The household-income value in thousands of euros.
- urban [
numeric] A trip-level count of how many of the trip's two endpoints (the origin and the destination) were classified as large cities in the original study:
0means that neither endpoint is a large city,1means that exactly one endpoint is a large city, and2means that both endpoints are large cities.
Source
Adapted from ModeCanada in the
mlogit package.
The source data were provided by Frank Koppelman and used by Bhat (1995).
References
Bhat CR (1995). “A heteroscedastic extreme value model of intercity travel mode choice.” Transportation Research Part B: Methodological, 29(6), 471–483. doi:10.1016/0191-2615(95)00015-6 .