Comparison of two non‐parametric models for daily traffic forecasting in Hong Kong

Abstract

The most up-to-date annual average daily traffic (AADT) is always required for transport model development and calibration. However, the current-year AADT data are not always available. The short-term traffic flow forecasting models can be used to predict the traffic flows for the current year. In this paper, two non-parametric models, non-parametric regression (NPR) and Gaussian maximum likelihood (GML), are chosen for short-term traffic forecasting based on historical data collected for the annual traffic census (ATC) in Hong Kong. These models are adapted as they are more flexible and efficient in forecasting the daily vehicular flows in the Hong Kong ATC core stations (in total of 87 stations). The daily vehicular flows predicted by these models are then used to calculate the AADT of the current year, 1999. The overall prediction and comparison results show that the NPR model produces better forecasts than the GML model using the ATC data in Hong Kong. Copyright © 2006 John Wiley _ Sons, Ltd.

Comparison of several traffic forecasting methods based on travel time index data on weekends

“…Considering the simplicity of the parametric techniques and the proved facts that nonparametric regression (NPR) performs better than the parametric counterpart [4][5] , we choose both parametric and nonparametric techniques in this work. Representative forecasting methods are investigated for the performance comparison as follows: historical-mean (HM), ARMA, Kalman filtering (KF), linear regression (LR), RBF networks and SVR models.…”

“…Lately extraordinary development of distinct nonparametric techniques, including nonparametric regression, neural networks, etc., has shown that their great potential alternative to their parametric counterparts. In essence, nonparametric statistical regression can be regarded as a dynamic clustering model that relies on the relationship between dependent and independent traffic variables [4][5] . In other words, it attempts to identify past information that is similar to the state at prediction time, and leads to easily implemented nature.…”

“…A large amount of scientific literatures have concerned empirical based predictions [4][5][6][7][8] . By applying statistical methodology or heuristic methods in traffic forecasting, the empirical approaches can be classified into two categories: parametric and nonparametric techniques.…”

Pedestrian Quantity Estimation with Trajectory Patterns

“…In the scenario of pedestrian quantity estimation, we directly model the edgeoriented quantities [5,6,15] within a Gaussian process regression framework. First, we convert the original networkG(Ṽ,Ẽ) to an edge graph G(V, E) that represents the adjacencies between edges ofG.…”

“…Improvements of this technique were achieved by respecting the geographical space by usage of geographical weighted regression (GWR) [4] and by application of k-nearest neighbor approaches (kNN) [5]. In [6] the AADT prediction of kNN for a particular location is improved by weighting measurements by their temporal distance to the prediction time. This approach showed better results than application of Gaussian maximum likelihood (GML) approaches for weighting of the historical data points.…”