biogeme 3.2.8 [2021-07-26]

Python package

Home page:

Submit questions to

Michel Bierlaire, Transport and Mobility Laboratory, Ecole Polytechnique Fédérale de Lausanne (EPFL)

This file has automatically been generated on 2021-07-26 19:10:15.944956

Report file: 21probit.html
Database name: swissmetro

Estimation report

Number of estimated parameters: 3
Sample size: 2232
Excluded observations: 8496
Init log likelihood: -1547.105
Final log likelihood: -986.1888
Likelihood ratio test for the init. model: 1121.831
Rho-square for the init. model: 0.363
Rho-square-bar for the init. model: 0.361
Akaike Information Criterion: 1978.378
Bayesian Information Criterion: 1995.51
Final gradient norm: 1.1085E-04
Nbr of threads: 36
Algorithm: Newton with trust region for simple bound constraints
Proportion analytical hessian: 100.0%
Relative projected gradient: 4.571628e-08
Relative change: 0.0005681240397018561
Number of iterations: 4
Number of function evaluations: 13
Number of gradient evaluations: 5
Number of hessian evaluations: 5
Cause of termination: Relative gradient = 4.6e-08 <= 6.1e-06
Optimization time: 0:00:00.017104

Estimated parameters

NameValueStd errt-testp-valueRob. Std errRob. t-testRob. p-value

Correlation of coefficients

Coefficient1Coefficient2CovarianceCorrelationt-testp-valueRob. cov.Rob. corr.Rob. t-testRob. p-value

Smallest eigenvalue: 321.491

Largest eigenvalue: 1145.47

Condition number: 3.563