Incorporating pro-environmental preferences towards green automobile technologies through a Bayesian hybrid choice model
Daziano, Bolduc, 2013, in Transportmetrica A: Transport Science
doi:10.1080/18128602.2010.524173
Location |
Canada |
Population |
General |
Sample size |
1877 |
Factor analysis type |
nan, nan rotation |
Stepwise regression |
no |
Removal of insignificant variables |
no |
Reviewed by |
2020-04-13 00:00:00 |
Abstract
Using stated data on both vehicle purchase decisions and environmental concerns, we develop, implement and apply the MCMC Gibbs sampler for Bayesian estimation of a Hybrid Choice Model (HCM). Whereas classical estimation of HCMs is fairly complex, we verify the feasibility of the Bayesian estimator as well as the HCM capacity to adapt to practical situations. We show that the Bayesian approach for HCMs is methodologically easier to implement than simulated maximum likelihood because the inclusion of latent variables translates into adding independent ordinary regressions; we also find that, using the Bayesian estimates, forecasting and deriving confidence intervals for willingness to pay measures is straightforward. Our empirical results coincide with a priori expectations, namely that environmentally-conscious consumers are willing to pay more for low-emission vehicles. The model outperforms standard discrete choice models because it not only incorporates pro-environmental preferences but also provides tools to build a profile of environmentally-conscious consumers.
Factors
Models
Dependent variable |
Type of car purchased |
Model type |
Bayesian ICLV |
Sample size |
1877.0 |
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 |
-1955.34 |
AFV |
Variable |
Coefficient |
p-value |
Alternative specific constant (AFV) |
-6.185
|
0.0 |
Environmental concern (Bayesian) |
0.585
|
0.0 |
Environmental concern |
Variable |
Coefficient |
p-value |
Intercept |
1.84
|
0.0 |
Driving alone user |
-0.157
|
0.026 |
Carpool user |
0.236
|
0.032 |
Transit user |
0.482
|
0.0 |
Female indicator |
0.344
|
0.0 |
High income indicator (> 80K$) |
0.046
|
0.441 |
University indicator |
0.274
|
0.0 |
Age level: 26-40 years |
0.447
|
0.0 |
Age level: 41-55 years |
0.544
|
0.0 |
Age level: 56 years and up |
0.839
|
0.0 |
HEV |
Variable |
Coefficient |
p-value |
Alternative specific constant (HEV) |
-2.53
|
0.0 |
Environmental concern (Bayesian) |
0.411
|
0.0 |
HFC |
Variable |
Coefficient |
p-value |
Alternative specific constant (HFC) |
-4.049
|
0.0 |
Environmental concern (Bayesian) |
0.674
|
0.0 |
Variable |
Coefficient |
p-value |
Purchase price |
-0.895
|
0.0 |
Fuel cost |
-0.852
|
0.0 |
Fuel availability |
1.388
|
0.0 |
Express lane access |
0.158
|
0.024 |
Power |
2.729
|
0.0 |
Dependent variable |
Type of car purchased |
Model type |
ICLV |
Sample size |
1877.0 |
R2 |
nan |
Adjusted R2 |
|
Pseudo R2
(nan)
|
nan |
AIC |
nan |
BIC |
nan |
Log-likelihood at zero |
nan |
Log-likelihood at constants |
nan |
Log-likelihood at convergence |
-1987.52 |
AFV |
Variable |
Coefficient |
p-value |
Alternative specific constant (AFV) |
-6.189
|
0.0 |
Environmental concern (classical) |
0.592
|
0.0 |
Environmental concern (classical) |
Variable |
Coefficient |
p-value |
Intercept |
2.067
|
0.0 |
Driving alone user |
-0.143
|
0.063 |
Carpool user |
0.241
|
0.086 |
Transit user |
0.468
|
0.0 |
Female indicator |
0.342
|
0.0 |
High income indicator (> 80K$) |
0.05
|
0.453 |
University indicator |
0.285
|
0.0 |
Age level: 26-40 years |
0.439
|
0.001 |
Age level: 41-55 years |
0.538
|
0.0 |
Age level: 56 years and up |
0.829
|
0.0 |
HEV |
Variable |
Coefficient |
p-value |
Alternative specific constant (HEV) |
-2.541
|
0.0 |
Environmental concern (classical) |
0.42
|
0.0 |
HFC |
Variable |
Coefficient |
p-value |
Alternative specific constant (HFC) |
-4.093
|
0.0 |
Environmental concern (classical) |
0.692
|
0.0 |
Variable |
Coefficient |
p-value |
Purchase price |
-0.894
|
0.0 |
Fuel cost |
-0.854
|
0.0 |
Fuel availability |
1.398
|
0.0 |
Express lane access |
0.16
|
0.024 |
Power |
2.752
|
0.0 |