Table 1. Model estimation results on HC and RC datasets, and the combined (RC+HC) data.
| Parameter | Estimate | ||||
|---|---|---|---|---|---|
| FP-MNL model | RP-MNL model | GMNL model | |||
| HC data | RC data | HC data | RC data | RC+HC data | |
| Means of utility coefficients | |||||
| DIST | -0.208 (-21.18)*** | -0.256 (-11.40)*** | -0.306 (-16.88)*** | -0.277 (-9.08)*** | -0.096 (-17.94)*** |
| CONG | -0.409 (-22.66)*** | -0.138 (-12.90)*** | -0.597 (-17.13)*** | -0.169 (-10.02)*** | -0.178 (-20.07)*** |
| FLTOVIS | -0.094 (-9.06)*** | -0.024 (-2.07)** | -0.119 (-7.22)*** | -0.035 (-2.37)** | -0.048 (-10.84)*** |
| FLTOINVIS | 0.054 (4.53)*** | 0.093 (9.52)*** | 0.055 (3.04)*** | 0.107 (7.75)*** | 0.047 (6.98)*** |
| VIS | 1.249 (12.14)*** | 0.710 (2.85)*** | 1.520 (10.06)*** | 0.828 (7.49)*** | 0.791 (15.26)*** |
| Standard deviations of coefficients | |||||
| DIST | 0.171 (9.90)*** | 0.091 (0.93) | 0.574 (4.02)*** | ||
| CONG | 0.327 (11.71)*** | 0.181 (4.48)*** | 0.325 (2.58)*** | ||
| FLTOVIS | 0.134 (9.27)*** | 0.070 (2.92)*** | 0.112 (0.79) | ||
| FLTOINVIS | 0.136 (9.30)*** | 0.072 (2.99)*** | 0.115 (0.80) | ||
| VIS | 1.190 (6.04)*** | 0.091 (0.42) | 0.068 (0.07) | ||
| Utility-scale dummy variable | 0.696 (12.48)*** | ||||
| Statistical fit measures | |||||
| Number of choice observations | 2338 | 3015 | 2338 | 3015 | 5353 |
| Initial log-likelihood | -3241.16 | -4174.13 | -3241.16 | -4174.13 | -7415.29 |
| Optimum log-likelihood | -2855.16 | -2863.54 | -2610.42 | -2854.74 | -5798.84 |
| Pseudo ρ2 | 0.119 | 0.314 | 0.195 | 0.316 | 0.218 |
Note that *** and ** indicate statistical significance at 99% and 95% level respectively. The values in parentheses show the t-statistic of the estimates.