This file has automatically been generated.
Wed Jul 6 20:03:14 2016
Tip: click on the columns headers to sort a table [Credits]
| Example of a logit model for a transportation mode choice with 3 alternatives: |
| - Train |
| - Car |
| - Swissmetro, an hypothetical high-speed train |
| We illustrate a case where persons from GROUP 3 have been undersampled. |
| In order to apply the WESML estimator, weights must be associated to observation. |
| A weight of 1.0 is associated with GROUP 2 and a weight of 1.2 with GROUP 3 |
| Note that the valie of the log likelihood cannot be compared with the ESML value. |
| Model: | Logit |
| Number of estimated parameters: | 4 |
| Number of observations: | 6768 |
| Number of individuals: | 6768 |
| Null log likelihood: | -7016.870 |
| Init log likelihood: | -7016.870 |
| Final log likelihood: | -5273.743 |
| Likelihood ratio test: | 3486.253 |
| Rho-square: | 0.248 |
| Adjusted rho-square: | 0.248 |
| Final gradient norm: | +6.154e-04 |
| Diagnostic: | Normal termination. Obj: 6.05545e-06 Const: 6.05545e-06 |
| Iterations: | 13 |
| Run time: | 00:00 |
| Variance-covariance: | from analytical hessian |
| Sample file: | ../swissmetro.dat |
| Name | Value | Std err | t-test | p-value | Robust Std err | Robust t-test | p-value | ||
|---|---|---|---|---|---|---|---|---|---|
| ASC_CAR | -0.114 | 0.0433 | -2.64 | 0.01 | 0.0590 | -1.94 | 0.05 | * | |
| ASC_SM | 0.00 | fixed | |||||||
| ASC_TRAIN | -0.757 | 0.0558 | -13.57 | 0.00 | 0.0824 | -9.18 | 0.00 | ||
| B_COST | -1.12 | 0.0522 | -21.43 | 0.00 | 0.0673 | -16.63 | 0.00 | ||
| B_TIME | -1.32 | 0.0576 | -22.94 | 0.00 | 0.106 | -12.50 | 0.00 |
| Id | Name | Availability | Specification |
|---|---|---|---|
| 1 | A1_TRAIN | TRAIN_AV_SP | ASC_TRAIN * one + B_TIME * TRAIN_TT_SCALED + B_COST * TRAIN_COST_SCALED |
| 2 | A2_SM | SM_AV | ASC_SM * one + B_TIME * SM_TT_SCALED + B_COST * SM_COST_SCALED |
| 3 | A3_Car | CAR_AV_SP | ASC_CAR * one + B_TIME * CAR_TT_SCALED + B_COST * CAR_CO_SCALED |
| Coefficient1 | Coefficient2 | Covariance | Correlation | t-test | p-value | Rob. cov. | Rob. corr. | Rob. t-test | p-value | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
| B_COST | B_TIME | 0.000589 | 0.196 | 2.89 | 0.00 | 0.00225 | 0.316 | 1.91 | 0.06 | * | |
| ASC_TRAIN | B_TIME | -0.00229 | -0.713 | 5.38 | 0.00 | -0.00773 | -0.887 | 3.09 | 0.00 | ||
| ASC_TRAIN | B_COST | -1.17e-05 | -0.00402 | 4.74 | 0.00 | -0.000900 | -0.162 | 3.17 | 0.00 | ||
| ASC_CAR | B_TIME | -0.00147 | -0.590 | 13.39 | 0.00 | -0.00497 | -0.796 | 7.70 | 0.00 | ||
| ASC_CAR | B_COST | 0.000466 | 0.206 | 16.60 | 0.00 | -4.15e-05 | -0.0104 | 11.17 | 0.00 | ||
| ASC_CAR | ASC_TRAIN | 0.00139 | 0.576 | 13.68 | 0.00 | 0.00394 | 0.810 | 13.11 | 0.00 |
Smallest singular value of the hessian: 5.89012