Citation

Pavlyuk D. Transp. Res. Proc. 2021; 52: 179-186.

Copyright

(Copyright © 2021, Elsevier Publications)

DOI

10.1016/j.trpro.2021.01.020

PMID

unavailable

Abstract

Spatiotemporal traffic forecasting models become a popular tool of urban transport engineering. Performance of spatiotemporal models and their generalisation abilities are the key aspects that are intensively addressed in methodological literature and case studies. This paper proposes a spatiotemporal cross-validation approach to estimating model performance, which extends classical temporal cross-validation techniques to a complex spatiotemporal structure of traffic flow relationships. The proposed approach allows estimating model generalisation abilities in the spatiotemporal dimension - ability to forecast traffic flows at unobserved nearby road segments. Additionally, the spatiotemporal cross-validation provides clues for stability of model performance with respect to minor modifications of the spatial structure. Advantages of the proposed spatiotemporal cross-validation approach are demonstrated on a large citywide traffic data set.

23rd EURO Working Group on Transportation Meeting, EWGT 2020, 16-18 September 2020, Paphos, Cyprus

Language: en

Keywords

machine learning; multivariate time series; short-term traffic forecasting; spatiotemporal structure