English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The mixed-logit model is a flexible tool in transportation choice analysis,which provides valuable insights into inter and intra-individual behaviouralheterogeneity. However, applications of mixed-logit models are limited by thehigh computational and data requirements for model estimation. When estimatingon small samples, the Bayesian estimation approach becomes vulnerable to overand under-fitting. This is problematic for investigating the behaviour ofspecific population sub-groups or market segments with low data availability.Similar challenges arise when transferring an existing model to a new locationor time period, e.g., when estimating post-pandemic travel behaviour. Wepropose an Early Stopping Bayesian Data Assimilation (ESBDA) simulator forestimation of mixed-logit which combines a Bayesian statistical approach withMachine Learning methodologies. The aim is to improve the transferability ofmixed-logit models and to enable the estimation of robust choice models withlow data availability. This approach can provide new insights into choicebehaviour where the traditional estimation of mixed-logit models was notpossible due to low data availability, and open up new opportunities forinvestment and planning decisions support. The ESBDA estimator is benchmarkedagainst the Direct Application approach, a basic Bayesian simulator with randomstarting parameter values and a Bayesian Data Assimilation (BDA) simulatorwithout early stopping. The ESBDA approach is found to effectively overcomeunder and over-fitting and non-convergence issues in simulation. Its resultingmodels clearly outperform those of the reference simulators in predictiveaccuracy. Furthermore, models estimated with ESBDA tend to be more robust, withsignificant parameters with signs and values consistent with behaviouraltheory, even when estimated on small samples.

An Early Stopping Bayesian Data Assimilation Approach for Mixed-Logit Estimation