Attitudes and value of time heterogeneity
Maya Abou-Zeid , Moshe Ben-Akiva , Michel Bierlaire , Charisma Choudhury , Stephane Hess, 2010, in Applied Transport Economics
doi:nan
Location |
Stockholm, Sweden |
Population |
Other (specify) |
Sample size |
2216 |
Factor analysis type |
confirmatory factor analysis, nan rotation |
Stepwise regression |
no |
Removal of insignificant variables |
nan |
Reviewed by |
MWC |
Abstract
There is ample evidence showing a high level of heterogeneity of values of time among travelers. Previous studies have represented this heterogeneity by a distribution such as lognormal whose parameters depend on covariates like income, trip purpose, and mode of travel. We present and demonstrate a model where the distribution of the value of time also depends on attitudes towards travel. Attitudes are latent, or unobservable, and their distribution determines the conditional distribution of the value of time given the observable covariates such as income. We illustrate this model using data from a stated preferences survey. The estimation results show that as expected the median value of time increases with income and that the variability of value of time also increases with income reflecting the greater effect that the attitude towards travel has for high income groups.
Factors
Models
Dependent variable |
Choice of car travel options, incl. number of speed cameras along route |
Model type |
multinomial logit |
Sample size |
2216.0 |
R2 |
nan |
Adjusted R2 |
|
Pseudo R2
(nan)
|
nan |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
-1575.566 |
Variable |
Coefficient |
p-value |
Alternative specific constant (first alternative) |
4.03
|
0.0 |
Alternative specific constant (second alternative) |
2.86
|
0.0 |
Travel time |
-0.0397
|
0.0 |
Cost / income |
-0.953
|
0.0 |
Number of speed cameras |
-0.11
|
0.006 |
Dependent variable |
Choice of car travel options, incl. number of speed cameras along route |
Model type |
integrated choice and latent variable |
Sample size |
2216.0 |
R2 |
nan |
Adjusted R2 |
|
Pseudo R2
(nan)
|
nan |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
-15266.435 |
Car-lover |
Variable |
Coefficient |
p-value |
Constant |
5.25
|
0.0 |
Female |
-0.0185
|
0.734 |
Income |
0.0347
|
0.046 |
Age < 55 |
-0.0217
|
0.066 |
Age 55-65 |
0.00797
|
0.381 |
Age > 65 |
0.0231
|
0.019 |
Education: primary |
-0.147
|
0.346 |
Education: university |
-0.252
|
0.0 |
Education: other |
-0.157
|
0.394 |
Std. dev. |
0.934
|
0.0 |
Utility |
Variable |
Coefficient |
p-value |
Alternative specific constant (first alternative) |
4.01
|
0.0 |
Alternative specific constant (second alternative) |
2.84
|
0.0 |
Travel time |
-0.0388
|
0.0 |
Cost / income |
-2.02
|
0.0 |
Cost / income * car lover |
0.265
|
0.036 |
Number of speed cameras |
-0.109
|
0.006 |