Model run by trashe using Apollo 0.4.0 on R 4.5.3 for Linux. Please acknowledge the use of Apollo by citing Hess & Palma (2019) DOI 10.1016/j.jocm.2019.100170 www.ApolloChoiceModelling.com Model name : ECL_preference_space_panel_effect Model description : Error components logit model on Swiss route choice data, uncorrelated Lognormals in preference space, with panel effect term Model run at : 2026-11-02 21:45:50 Estimation method : bgw Estimation diagnosis : Relative function convergence (favorable) Optimisation diagnosis : Maximum found hessian properties : Negative definite maximum eigenvalue : -34.867321 reciprocal of condition number : 0.0564265 Number of individuals : 388 Number of rows in database : 3492 Number of modelled outcomes : 3492 Number of cores used : 4 Number of inter-individual draws : 500 (mlhs) LL(start) : -2253.75 LL at equal shares, LL(0) : -2420.47 LL at observed shares, LL(C) : -2420.39 LL(final) : -1440.55 Rho-squared vs equal shares : 0.4048 Adj.Rho-squared vs equal shares : 0.4011 Rho-squared vs observed shares : 0.4048 Adj.Rho-squared vs observed shares : 0.4015 AIC : 2899.11 BIC : 2954.53 Estimated parameters : 9 Time taken (hh:mm:ss) : 00:00:26.36 pre-estimation : 00:00:4.49 estimation : 00:00:9.21 post-estimation : 00:00:12.66 Iterations : 19 Unconstrained optimisation. Estimates: Estimate s.e. t.rat.(0) Rob.s.e. Rob.t.rat.(0) mu_log_b_tt -1.9494 0.09159 -21.285 0.12295 -15.855 sigma_log_b_tt 0.4752 0.07931 5.992 0.08568 5.547 mu_log_b_tc -1.0174 0.13457 -7.560 0.17271 -5.891 sigma_log_b_tc 1.0293 0.09390 10.961 0.10029 10.263 mu_log_b_hw -2.8876 0.08477 -34.064 0.08838 -32.675 sigma_log_b_hw 0.7613 0.10206 7.459 0.10439 7.293 mu_log_b_ch 0.6804 0.07808 8.714 0.08538 7.970 sigma_log_b_ch -0.8635 0.10288 -8.393 0.11343 -7.612 sigma_panel 0.3933 0.08943 4.398 0.10642 3.696 Overview of choices for MNL model component : alt1 alt2 Times available 3492.00 3492.00 Times chosen 1734.00 1758.00 Percentage chosen overall 49.66 50.34 Percentage chosen when available 49.66 50.34 Classical covariance matrix: mu_log_b_tt sigma_log_b_tt mu_log_b_tc sigma_log_b_tc mu_log_b_tt 0.008388 -0.002774 0.008761 -0.002461 sigma_log_b_tt -0.002774 0.006290 -0.001251 0.001110 mu_log_b_tc 0.008761 -0.001251 0.018110 -0.007627 sigma_log_b_tc -0.002461 0.001110 -0.007627 0.008818 mu_log_b_hw 0.002035 8.2959e-04 0.002069 6.0031e-04 sigma_log_b_hw 9.4827e-04 -4.8631e-04 7.9823e-04 -1.3747e-04 mu_log_b_ch 0.002561 3.3876e-04 0.002626 4.9821e-04 sigma_log_b_ch -0.001733 5.5300e-04 -0.002501 -8.552e-05 sigma_panel 0.001826 3.0061e-04 0.001857 3.5003e-04 mu_log_b_hw sigma_log_b_hw mu_log_b_ch sigma_log_b_ch mu_log_b_tt 0.002035 9.4827e-04 0.002561 -0.001733 sigma_log_b_tt 8.2959e-04 -4.8631e-04 3.3876e-04 5.5300e-04 mu_log_b_tc 0.002069 7.9823e-04 0.002626 -0.002501 sigma_log_b_tc 6.0031e-04 -1.3747e-04 4.9821e-04 -8.552e-05 mu_log_b_hw 0.007186 -0.002523 0.001796 -7.8108e-04 sigma_log_b_hw -0.002523 0.010417 0.001209 -2.4379e-04 mu_log_b_ch 0.001796 0.001209 0.006096 -6.5776e-04 sigma_log_b_ch -7.8108e-04 -2.4379e-04 -6.5776e-04 0.010583 sigma_panel 0.001926 -3.5672e-04 0.001984 -4.2061e-04 sigma_panel mu_log_b_tt 0.001826 sigma_log_b_tt 3.0061e-04 mu_log_b_tc 0.001857 sigma_log_b_tc 3.5003e-04 mu_log_b_hw 0.001926 sigma_log_b_hw -3.5672e-04 mu_log_b_ch 0.001984 sigma_log_b_ch -4.2061e-04 sigma_panel 0.007997 Robust covariance matrix: mu_log_b_tt sigma_log_b_tt mu_log_b_tc sigma_log_b_tc mu_log_b_tt 0.015117 -0.005380 0.018597 -0.006068 sigma_log_b_tt -0.005380 0.007341 -0.005805 0.003811 mu_log_b_tc 0.018597 -0.005805 0.029829 -0.011694 sigma_log_b_tc -0.006068 0.003811 -0.011694 0.010058 mu_log_b_hw 0.003545 5.4497e-04 0.003911 6.2560e-04 sigma_log_b_hw 0.001113 -9.1109e-04 0.001360 -8.7226e-04 mu_log_b_ch 0.004760 -2.2286e-04 0.006070 -5.2955e-04 sigma_log_b_ch -0.002545 6.9771e-04 -0.004593 1.4229e-04 sigma_panel 0.002134 6.659e-05 0.002129 4.4856e-04 mu_log_b_hw sigma_log_b_hw mu_log_b_ch sigma_log_b_ch mu_log_b_tt 0.003545 0.001113 0.004760 -0.002545 sigma_log_b_tt 5.4497e-04 -9.1109e-04 -2.2286e-04 6.9771e-04 mu_log_b_tc 0.003911 0.001360 0.006070 -0.004593 sigma_log_b_tc 6.2560e-04 -8.7226e-04 -5.2955e-04 1.4229e-04 mu_log_b_hw 0.007810 -0.002749 0.003371 -6.2871e-04 sigma_log_b_hw -0.002749 0.010897 8.5682e-04 -0.001278 mu_log_b_ch 0.003371 8.5682e-04 0.007289 -0.001581 sigma_log_b_ch -6.2871e-04 -0.001278 -0.001581 0.012866 sigma_panel 0.001318 5.9356e-04 0.002128 -0.001647 sigma_panel mu_log_b_tt 0.002134 sigma_log_b_tt 6.659e-05 mu_log_b_tc 0.002129 sigma_log_b_tc 4.4856e-04 mu_log_b_hw 0.001318 sigma_log_b_hw 5.9356e-04 mu_log_b_ch 0.002128 sigma_log_b_ch -0.001647 sigma_panel 0.011326 Classical correlation matrix: mu_log_b_tt sigma_log_b_tt mu_log_b_tc sigma_log_b_tc mu_log_b_tt 1.0000 -0.38194 0.71080 -0.286159 sigma_log_b_tt -0.3819 1.00000 -0.11719 0.148986 mu_log_b_tc 0.7108 -0.11719 1.00000 -0.603595 sigma_log_b_tc -0.2862 0.14899 -0.60360 1.000000 mu_log_b_hw 0.2622 0.12339 0.18140 0.075415 sigma_log_b_hw 0.1014 -0.06008 0.05812 -0.014344 mu_log_b_ch 0.3581 0.05470 0.24990 0.067953 sigma_log_b_ch -0.1839 0.06778 -0.18067 -0.008853 sigma_panel 0.2230 0.04238 0.15427 0.041683 mu_log_b_hw sigma_log_b_hw mu_log_b_ch sigma_log_b_ch mu_log_b_tt 0.26217 0.10145 0.35813 -0.183879 sigma_log_b_tt 0.12339 -0.06008 0.05470 0.067777 mu_log_b_tc 0.18140 0.05812 0.24990 -0.180668 sigma_log_b_tc 0.07542 -0.01434 0.06795 -0.008853 mu_log_b_hw 1.00000 -0.29166 0.27133 -0.089566 sigma_log_b_hw -0.29166 1.00000 0.15174 -0.023219 mu_log_b_ch 0.27133 0.15174 1.00000 -0.081887 sigma_log_b_ch -0.08957 -0.02322 -0.08189 1.000000 sigma_panel 0.25403 -0.03908 0.28409 -0.045719 sigma_panel mu_log_b_tt 0.22299 sigma_log_b_tt 0.04238 mu_log_b_tc 0.15427 sigma_log_b_tc 0.04168 mu_log_b_hw 0.25403 sigma_log_b_hw -0.03908 mu_log_b_ch 0.28409 sigma_log_b_ch -0.04572 sigma_panel 1.00000 Robust correlation matrix: mu_log_b_tt sigma_log_b_tt mu_log_b_tc sigma_log_b_tc mu_log_b_tt 1.00000 -0.510688 0.87579 -0.49214 sigma_log_b_tt -0.51069 1.000000 -0.39229 0.44352 mu_log_b_tc 0.87579 -0.392294 1.00000 -0.67511 sigma_log_b_tc -0.49214 0.443517 -0.67511 1.00000 mu_log_b_hw 0.32623 0.071974 0.25620 0.07058 sigma_log_b_hw 0.08670 -0.101871 0.07541 -0.08332 mu_log_b_ch 0.45344 -0.030467 0.41168 -0.06185 sigma_log_b_ch -0.18245 0.071795 -0.23445 0.01251 sigma_panel 0.16306 0.007304 0.11581 0.04203 mu_log_b_hw sigma_log_b_hw mu_log_b_ch sigma_log_b_ch mu_log_b_tt 0.32623 0.08670 0.45344 -0.18245 sigma_log_b_tt 0.07197 -0.10187 -0.03047 0.07179 mu_log_b_tc 0.25620 0.07541 0.41168 -0.23445 sigma_log_b_tc 0.07058 -0.08332 -0.06185 0.01251 mu_log_b_hw 1.00000 -0.29798 0.44673 -0.06272 sigma_log_b_hw -0.29798 1.00000 0.09614 -0.10796 mu_log_b_ch 0.44673 0.09614 1.00000 -0.16330 sigma_log_b_ch -0.06272 -0.10796 -0.16330 1.00000 sigma_panel 0.14016 0.05343 0.23417 -0.13642 sigma_panel mu_log_b_tt 0.163063 sigma_log_b_tt 0.007304 mu_log_b_tc 0.115815 sigma_log_b_tc 0.042027 mu_log_b_hw 0.140159 sigma_log_b_hw 0.053429 mu_log_b_ch 0.234167 sigma_log_b_ch -0.136424 sigma_panel 1.000000 Settings and functions used in model definition: apollo_control -------------- Value modelName "ECL_preference_space_panel_effect" modelDescr "Error components logit model on Swiss route choice data, uncorrelated Lognormals in preference space, with panel effect term" indivID "ID" nCores "4" outputDirectory "output/" mixing "TRUE" debug "FALSE" workInLogs "FALSE" seed "13" HB "FALSE" noValidation "FALSE" noDiagnostics "FALSE" calculateLLC "TRUE" analyticHessian "FALSE" dropLoopExpansion "FALSE" dropRedefCheck "FALSE" panelData "TRUE" analyticGrad "TRUE" analyticGrad_manualSet "FALSE" overridePanel "FALSE" preventOverridePanel "FALSE" noModification "FALSE" Hessian routines attempted -------------------------- numerical jacobian of LL analytical gradient apollo_randCoeff ------------------ function (apollo_beta, apollo_inputs) { randcoeff = list() randcoeff[["b_tt"]] = -exp(mu_log_b_tt + sigma_log_b_tt * draws_tt) randcoeff[["b_tc"]] = -exp(mu_log_b_tc + sigma_log_b_tc * draws_tc) randcoeff[["b_hw"]] = -exp(mu_log_b_hw + sigma_log_b_hw * draws_hw) randcoeff[["b_ch"]] = -exp(mu_log_b_ch + sigma_log_b_ch * draws_ch) randcoeff[["ec_alt1"]] = sigma_panel * draws_alt1 randcoeff[["ec_alt2"]] = sigma_panel * draws_alt2 return(randcoeff) } apollo_probabilities ---------------------- function (apollo_beta, apollo_inputs, functionality = "estimate") { apollo_attach(apollo_beta, apollo_inputs) on.exit(apollo_detach(apollo_beta, apollo_inputs)) P = list() V = list() V[["alt1"]] = b_tt * tt1 + b_tc * tc1 + b_hw * hw1 + b_ch * ch1 + ec_alt1 V[["alt2"]] = b_tt * tt2 + b_tc * tc2 + b_hw * hw2 + b_ch * ch2 + ec_alt2 mnl_settings = list(alternatives = c(alt1 = 1, alt2 = 2), avail = list(alt1 = 1, alt2 = 1), choiceVar = choice, utilities = V) P[["model"]] = apollo_mnl(mnl_settings, functionality) P = apollo_panelProd(P, apollo_inputs, functionality) P = apollo_avgInterDraws(P, apollo_inputs, functionality) P = apollo_prepareProb(P, apollo_inputs, functionality) return(P) }