The estimation of generalized extreme value models from choice-based samples

Sampling issues are of greater importance in the estimation of statistical models in general, and of discrete choice models in particular. Although simple random sampling strategies are theoretically convenient, they are practically never feasible in a transportation context. However, it is well known that maximum likelihood estimation of the multinomial logit (MNL) model on stratified samples yields to consistent estimates for all parameters except the constants, which must be corrected afterward. This property does not generalize to non-MNL generalized extreme value (GEV) models, so that classical estimation procedures do not yield to maximum likelihood estimators from choice-based samples. In this paper, we propose a simple estimator providing consistent estimates of all parameters, including the constants, for non-MNL GEV models, which does not require an a priori knowledge of the sampling probabilities. The new estimator require minor modifications of existing estimation codes. We illustrate the concept on synthetic and real data for nested and cross-nested logit models. The results show that the classical "exogenous sample maximum likelihood" estimator produces biased estimates, even for parameters other than the constants. The new estimator is able to identify all the parameters, including the constants if the GEV model is non-trivial. For the covering abstract see ITRD E135582.

• Record URL:
  • http://abstracts.aetransport.org/conference/index/id/12
• Availability:
  • Find a library where document is available. Order URL: http://etcproceedings.org/conference
• Corporate Authors:

  Association for European Transport (AET)

  1 Vernon Mews, Vernon Street, West Kensington
  London W14 0RL,    
• Authors:
  • BIERLAIRE, M
  • BOLDUC, D
  • McFadden, D
• Conference:
  • European Transport Conference (ETC)
  • Location: Strasbourg , France
  • Date: 2006-9-0 to 2006-9-0
• Publication Date: 2006

Language

• English

Media Info

• Pagination: 24p
• Monograph Title: Proceedings of the European Transport Conference (ETC) 2006, held September 2006, Strasbourg, France

Subject/Index Terms

• TRT Terms: Conferences; Evaluation; Forecasting; Mathematical analysis; Mathematical models
• Uncontrolled Terms: Selection
• ITRD Terms: 6471: Analysis (math); 8525: Conference; 9020: Evaluation (assessment); 132: Forecast; 6473: Mathematical model; 9072: Selection
• Subject Areas: Operations and Traffic Management; I71: Traffic Theory;

Filing Info

• Accession Number: 01087210
• Record Type: Publication
• Source Agency: Transport Research Laboratory
• ISBN: 1905701012
• Files: ITRD
• Created Date: Jan 29 2008 9:57AM