A hybrid linear-neural model for time series forecasting
Authors:
- A. VeigaAVA. Veiga
- M.C. MedeirosMMM.C. Medeiros
- 1 November 2000
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks
- Vol. 11 (6) , 1402-1412
Abstract
This paper considers a linear model with time varying parameters controlled by a neural network to analyze and forecast nonlinear time series. We show that this formulation, called neural coefficient smooth transition autoregressive model, is in close relation to the threshold autoregressive model and the smooth transition autoregressive model with the advantage of naturally incorporating linear multivariate thresholds and smooth transitions between regimes. In our proposal, the neural-network output is used to induce a partition of the input space, with smooth and multivariate thresholds. This also allows the choice of good initial values for the training algorithm.Keywords
This publication has 19 references indexed in Scilit:
- A hybrid linear-neural model for time series forecastingIEEE Transactions on Neural Networks, 2000
- Modeling Multiple Regimes in the Business CycleMacroeconomic Dynamics, 1999
- Forecasting with artificial neural networks:International Journal of Forecasting, 1998
- Unit roots and smooth transitionsJournal of Time Series Analysis, 1998
- Specification, Estimation, and Evaluation of Smooth Transition Autoregressive ModelsJournal of the American Statistical Association, 1994
- Identification of fuzzy prediction models through hyperellipsoidal clusteringIEEE Transactions on Systems, Man, and Cybernetics, 1994
- Testing and Modeling Threshold Autoregressive ProcessesJournal of the American Statistical Association, 1989
- ON ESTIMATING THRESHOLDS IN AUTOREGRESSIVE MODELSJournal of Time Series Analysis, 1986
- FORECASTING TIME SERIES:A COMPARATIVE ANALYSIS OF ALTERNATIVE CLASSES OF TIME SERIES MODELSJournal of Time Series Analysis, 1985
- Modelling Nonlinear Random Vibrations Using an Amplitude-Dependent Autoregressive Time Series ModelBiometrika, 1981