Dynamical functional prediction and classification, with application to traffic flow prediction

TL;DR

This paper introduces a functional data analysis method that predicts and classifies traffic flow patterns by modeling trajectories as mixtures of stochastic processes, enabling accurate forecasting and pattern identification.

Contribution

It presents a novel functional mixture prediction approach combining probabilistic classification and clustering for traffic flow analysis, addressing pattern variability.

Findings

Effective prediction of traffic flow trajectories.

Identification of distinct traffic flow patterns.

Applicable to various longitudinal functional data.

Abstract

Motivated by the need for accurate traffic flow prediction in transportation management, we propose a functional data method to analyze traffic flow patterns and predict future traffic flow. In this study we approach the problem by sampling traffic flow trajectories from a mixture of stochastic processes. The proposed functional mixture prediction approach combines functional prediction with probabilistic functional classification to take distinct traffic flow patterns into account. The probabilistic classification procedure, which incorporates functional clustering and discrimination, hinges on subspace projection. The proposed methods not only assist in predicting traffic flow trajectories, but also identify distinct patterns in daily traffic flow of typical temporal trends and variabilities. The proposed methodology is widely applicable in analysis and prediction of longitudinally recorded functional data.