Data on travel mode choice for travel between Sydney and Melbourne, Australia.
Usage
data("TravelMode")
Format
A data frame containing 840 observations on 4 modes for 210 individuals.
This corresponds to so-called long format since we have 4 rows per
individual in a situation with 4 classes/categories of the response variable mode.
individual
Factor indicating individual with levels 1 to 200.
mode
Factor indicating travel mode with levels
"car", "air", "train", or "bus".
choice
Factor indicating choice with levels "no" and "yes".
wait
Terminal waiting time, 0 for car.
vcost
Vehicle cost component.
travel
Travel time in the vehicle.
gcost
Generalized cost measure.
income
Household income.
size
Party size.
Details
Data and description are taken from R package AER by Christian Kleiber and Achim Zeileis.
In MRSP-terminology, variables wait, vcost, travel and gcost are
category-specific since they take different values for different categories of mode. The variables
income and size are global. In the literature on discrete choice modelling,
these two variable types are called “alternative-specific” and “individual-specific”, respectively.
Source
R package AER, which states its source as follows: