Evaluation of Optimization Methods for Estimating Mixed Logit Models

The performances of different simulation-based estimation techniques for mixed logit modeling are evaluated. A quasi-Monte Carlo method (modified Latin hypercube sampling) is compared with a Monte Carlo algorithm with dynamic accuracy. The classic Broyden-Fletcher-Goldfarb-Shanno (BFGS) optimization algorithm line-search approach and trust region methods, which have proved to be extremely powerful in nonlinear programming, are also compared. Numerical tests are performed on two real data sets: stated preference data for parking type collected in the United Kingdom, and revealed preference data for mode choice collected as part of a German travel diary survey. Several criteria are used to evaluate the approximation quality of the log likelihood function and the accuracy of the results and the associated estimation runtime. Results suggest that the trust region approach outperforms the BFGS approach and that Monte Carlo methods remain competitive with quasi-Monte Carlo methods in high-dimensional problems, especially when an adaptive optimization algorithm is used.

[1]  K Axhausen,et al.  THE BIRMINGHAM CLAMP STATED PREFERENCE SURVEY. 2ND INTERIM REPORT TO BIRMINGHAM CITY COUNCIL , 1989 .

[2]  Kay W. Axhausen,et al.  The Birmingham CLAMP Stated Preference Survey , 1989 .

[3]  Kay W. Axhausen,et al.  The application of CLAMP to the analysis of parking policy in Birmingham City centre , 1990 .

[4]  David J. Thuente,et al.  Line search algorithms with guaranteed sufficient decrease , 1994, TOMS.

[5]  Lung-fei Lee,et al.  Asymptotic Bias in Simulated Maximum Likelihood Estimation of Discrete Choice Models , 1995, Econometric Theory.

[6]  Jorge Nocedal,et al.  Algorithm 778: L-BFGS-B: Fortran subroutines for large-scale bound-constrained optimization , 1997, TOMS.

[7]  K. Train,et al.  Mixed Logit with Repeated Choices: Households' Choices of Appliance Efficiency Level , 1998, Review of Economics and Statistics.

[8]  Stephen J. Wright,et al.  Numerical Optimization , 2018, Fundamental Statistical Inference.

[9]  C. Bhat.  Quasi-random maximum simulated likelihood estimation of the mixed multinomial logit model , 2001 .

[10]  Chandra R. Bhat,et al.  A comprehensive daily activity-travel generation model system for workers , 2000 .

[11]  Joffre Swait,et al.  Distinguishing taste variation from error structure in discrete choice data , 2000 .

[12]  K. Axhausen,et al.  Observing the rhythms of daily life , 2000 .

[13]  K. Train.  Halton Sequences for Mixed Logit , 2000 .

[14]  D. McFadden,et al.  MIXED MNL MODELS FOR DISCRETE RESPONSE , 2000 .

[15]  Nicholas I. M. Gould,et al.  Trust Region Methods , 2000, MOS-SIAM Series on Optimization.

[16]  Joan L. Walker.  Extended discrete choice models : integrated framework, flexible error structures, and latent variables , 2001 .

[17]  Pierre L'Ecuyer,et al.  Recent Advances in Randomized Quasi-Monte Carlo Methods , 2002 .

[18]  C. Bhat.  Simulation estimation of mixed discrete choice models using randomized and scrambled Halton sequences , 2003 .

[19]  A. Daly,et al.  On the performance of the shuffled Halton sequence in the estimation of discrete choice models , 2003 .

[20]  Kenneth E. Train,et al.  Discrete Choice Methods with Simulation , 2016 .

[21]  R. Garrido.  Estimation Performance of Low Discrepancy Sequences in Stated Preferences , 2003 .

[22]  Zsolt Sándor,et al.  Quasi-random simulation of discrete choice models , 2004 .

[23]  Kay W. Axhausen,et al.  Evidence on the distribution of values of travel time savings from a six-week diary , 2004 .

[24]  M. Bierlaire,et al.  ESTIMATION OF VALUE OF TRAVEL-TIME SAVINGS USING MIXED LOGIT MODELS , 2005 .

[25]  Philippe L. Toint,et al.  An adaptive Monte Carlo algorithm for computing mixed logit estimators , 2006, Comput. Manag. Sci..

[26]  Philippe L. Toint,et al.  Convergence theory for nonconvex stochastic programming with an application to mixed logit , 2006, Math. Program..

[27]  Stephane Hess,et al.  On the use of a Modified Latin Hypercube Sampling (MLHS) method in the estimation of a Mixed Logit Model for vehicle choice , 2006 .