Bayesian Disaggregate and Aggregate Calibration of Path Logit Choice Models

Table 2

Calibration results for the path choice model, aggregate data.

ID, approach	5, C	6, C	7, B	8, B
Interviews	124	186	124	186
θ T ∈ ]0, +∞[(1/h): travel time	−1.72	−1.76	−1.13	−1.18
θ P ∈ [0, 1]: percentage on motorway	0.25	0.24	0.92	0.91
θ L ∈ {0, 1}: label time	1.65	1.72	2.06	2.14

C, Classic; B, Bayesian; ln (likelihood): with parameters equal to zero = −204.34; with best parameters = −148.85; pseudo-ρ² = 0.272.