Prediction Intervals for a Noisy Nonlinear Time Series Based on a Bootstrapping Reservoir Computing Network Ensemble
Authors:
- 14 March 2013
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 24 (7) , 1036-1048
Abstract
Prediction intervals that provide estimated values as well as the corresponding reliability are applied to nonlinear time series forecast. However, constructing reliable prediction intervals for noisy time series is still a challenge. In this paper, a bootstrapping reservoir computing network ensemble (BRCNE) is proposed and a simultaneous training method based on Bayesian linear regression is developed. In addition, the structural parameters of the BRCNE, that is, the number of reservoir computing networks and the reservoir dimension, are determined off-line by the 0.632 bootstrap cross-validation. To verify the effectiveness of the proposed method, two kinds of time series data, including the multisuperimposed oscillator problem with additive noises and a practical gas flow in steel industry are employed here. The experimental results indicate that the proposed approach has a satisfactory performance on prediction intervals for practical applications.Keywords
This publication has 26 references indexed in Scilit:
- On the use of cross-validation for time series predictor evaluationInformation Sciences, 2012
- A survey of cross-validation procedures for model selectionStatistics Surveys, 2010
- Decoupled echo state networks with lateral inhibitionNeural Networks, 2007
- Harnessing Nonlinearity: Predicting Chaotic Systems and Saving Energy in Wireless CommunicationScience, 2004
- Evaluation of several nonparametric bootstrap methods to estimate confidence intervals for software metricsIEEE Transactions on Software Engineering, 2003
- A constructive algorithm for training cooperative neural network ensemblesIEEE Transactions on Neural Networks, 2003
- Energy clearing price prediction and confidence interval estimation with cascaded neural networksIEEE Transactions on Power Systems, 2003
- Confidence estimation methods for neural networks: a practical comparisonIEEE Transactions on Neural Networks, 2001
- Bayesian approach to neural-network modeling with input uncertaintyIEEE Transactions on Neural Networks, 1999
- Prediction Intervals for Artificial Neural NetworksJournal of the American Statistical Association, 1997