Titheridge, H; Liu, C-P; (2016) Applying a Multilevel MNL Model to Analyse the Land Use Influence on Mode Choice Behaviour in Taiwan. In: Proceedings of the 95th TRB Annual Meeting. TRB: Washington DC, USA.

Preview

Text CPL HT TRB_2016-Nov15th_revised_submitted.pdf - Accepted Version Download (7MB) | Preview

Abstract

This study introduced a multilevel MNL model to explore the unobserved spatial heterogeneity 3 and the impact of land use and public transport provision at district and city/county levels on 4 mode choice between car, motorbike and public transport in Taiwan. The study found that the 5 unobserved spatial heterogeneity does exert significant effects on mode choice behaviour. In 6 addition, by comparing the results from a single-level MNL model and the multilevel MNL 7 model, it was shown that the multilevel MNL model has a better fit but also provides evidence 8 that neglecting the spatial autocorrelation and spatial heterogeneity could create misleading 9 results. This study also found that, in Taiwan, higher population density at the district and 10 city/county levels is associated with a higher probability of choosing public transport over the 11 car and motorbike. However, more diversified land uses and more grid-like street patterns are 12 associated with a higher probability of motorbike use. These results may contribute to planning 13 motorbike management strategies, amongst others, for Southeast Asian countries.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as XMLJSONCSV

Archive Staff Only

View Item