Dynamical functional prediction and classification, with application to traffic flow prediction
Journal
The Annals of Applied Statistics
Journal Volume
6
Journal Issue
4
ISSN
1932-6157
Date Issued
2012-12-01
Author(s)
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. © Institute of Mathematical Statistics, 2012.
Subjects
Clustering
Discrimination
Functional regression
Intelligent transportation system
Mixture model
Subspace projection
Traffic flow rate
Publisher
Institute of Mathematical Statistics
Type
journal article