Comparison of parametric and nonparametric models for traffic flow forecasting

By Brian L. Smith, Billy M. Williams, R. Oswald. 2002

983 citations recorded by the source corpus.

Subfields: Building and Construction, Environmental Engineering, Transportation

View Paper: https://www.semanticscholar.org/paper/1a896e5c49ade276b5e03f952b29b2df93d21129

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Building and Construction

In Environmental Engineering

In Transportation

Citations in the corpus

Online-SVR for short-term traffic flow prediction under typical and atypical traffic conditions (2009, methodology): “Several other techniques have been applied to predict real-time traffic flow, including multivariate state space time series (Stathopoulos & Karlaftis, 2003), multivariate non-parametric regression (Clark, 2003; Smith & Demetsky, 1996), nearest neighbor non-parametric regression (Davis & Nihan, 1991”
Long short-term memory neural network for traffic speed prediction using remote microwave sensor data (2015, background): “In the past decades, a number of ARIMA based time series models have been proposed for traffic prediction (Williams and Hoel, 2003; Smith et al., 2002; Williams, 2001; Chandra and Al-Deek, 2009).”
A novel wavelet-SVM short-time passenger flow prediction in Beijing subway system (2015, methodology): “The nearest neighboring analysis model regards the predicted traffic data as non-given points in a certain space, and interpolates them with the historical traffic data points [10,12].”
Spatial temporal incidence dynamic graph neural networks for traffic flow forecasting (2020, methodology): “Generally, traditional traffic passenger flows prediction methods can be categorized into classic statistical models, including Auto-Regressive Integrated Moving Average (ARIMA) based methods [25, 26] and parametric learning methods, including K-nearest neighbor (KNN) nonparametric regression method”
Flow Prediction in Spatio-Temporal Networks Based on Multitask Deep Learning (2020, methodology): “Existing timeseries models, like the auto-regressive integrated moving average model (ARIMA, [3]), seasonal ARIMA [26], and the vector autoregressive model [4] can capture the temporal dependencies very well, yet it fails to handle spatial correlations.”
Short-term traffic flow prediction using seasonal ARIMA model with limited input data (2015, methodology): “In most of the studies on flow prediction [23, 24, 29, 30, 35], a MAPE in the range of 10–20 % was reported.”
Short‐term traffic forecasting: Overview of objectives and methods (2004, methodology): “Finally, Smith et al. (2002) tested the performance of non-parametric regression based on heuristically improved forecast generation methods and found that this approach did not produce better predictions than seasonal ARIMA.”
Traffic Flow Prediction With Big Data: A Deep Learning Approach (2015, methodology): “The SARIMA models and the nonparametric regression forecasting methods were evaluated in [49].”
Road Traffic Forecasting: Recent Advances and New Challenges (2018, background): “Although Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) are usual metrics for model performance measuring [10], [11], [42], they do not provide comparable measures when the complexities of compared models are too divergent [8]—e.”
Forecasting the short-term metro passenger flow with empirical mode decomposition and neural networks (2012, methodology): “The short-term transportation forecasting approaches can be generally divided into two categories: parametric and non-parametric techniques (Smith et al., 2002; Vlahogianni et al., 2004).”
An Aggregation Approach to Short-Term Traffic Flow Prediction (2009, background): “[21] J. M. Bates and C. W. J. Granger, “The combination of forecasts,” Oper.”