Modelling the Air Ticket Purchase Behavior Incorporating Latent Class Model

Table 4

Results of the nested logit model.

Variable	Coefficient
Peak-hour flight	3.162 (0.000)
Normal day-time flight	3.245 (0.000)
Evening flight	3.076 (0.065)
Normal train	1.890 (0.158)
High-speed train	2.764 (0.076)
Private car	0.561 (0.111)
Rented car	0.874 (0.107)
Daytime coach	0.145 (0.000)
Age	−0.231 (0.000)
Occupation	0.086 (0.354)
Position	0.149 (0.027)
Income	0.017 (0.216)
Frequent travel purpose	0.159 (0.000)
Long-distance frequency	−0.356 (0.000)
Purchase date for business	−0.002 (0.218
Purchase date for leisure	−0.016 (0.219)
Origin	0.001 (3.654)
Destination	−0.001 (1.769)
Travel purpose	0.032 (0.000)
Travel distance	−0.005 (0.119)
Departure time	−0.184 (0.379)
Travel time	−1.017 (0.062)
Delay time	−0.514 (0.000)
Access time	−0.043 (0.650)
Air fare	−1.015 (0.000)
Normal train fare	−0.547 (0.095)
High-speed train fare	−0.972 (0.000)
Rented car fare	−0.867 (0.821)
Private car fare	−0.708 (0.104)
Coach fare	−0.559 (2.732)
Loglikelihood function: −2596.37
Pseudo R^2 (ρ^2): 0.35

Significant at the 1% level. Significant at the 5% level. Significant at the 10% level.