This project is motivated by the possible value of integrating theory-based discrete choice models (DCMs) and data-driven neural networks. How to benefit from the strengths of both is the overarching question. We propose hybrid structures and strategies to flexibly represent taste heterogeneity and improve predictability while keeping model interpretability. Also, we utilize neural networks’ training machinery to speed up and scale up the estimation of Latent Class Choice Models (LCCMs).
Airport choice is an important air travel-related decision in multiple airport regions. This report proposes the use of a probabilistic choice set mul...
Links with this icon indicate that you are leaving a Bureau of Transportation
Statistics (BTS)/National Transportation Library (NTL)
Web-based service.
Thank you for visiting.
You are about to access a non-government link outside of
the U.S. Department of Transportation's National
Transportation Library.
Please note: While links to Web sites outside of DOT are
offered for your convenience, when you exit DOT Web sites,
Federal privacy policy and Section 508 of the Rehabilitation
Act (accessibility requirements) no longer apply. In
addition, DOT does not attest to the accuracy, relevance,
timeliness or completeness of information provided by linked
sites. Linking to a Web site does not constitute an
endorsement by DOT of the sponsors of the site or the
products presented on the site. For more information, please
view DOT's Web site linking policy.
To get back to the page you were previously viewing, click
your Cancel button.