A novel neural network ensemble architecture for time series forecasting

Authors:
Iffat A. Gheyas;Leslie S. Smith
Affiliations:
University of Aberdeen Business School, Edward Wright Building, Dunbar Street, Old Aberdeen AB24 3QY, United Kingdom;University of Aberdeen Business School, Edward Wright Building, Dunbar Street, Old Aberdeen AB24 3QY, United Kingdom
Venue:
Neurocomputing
Year:
2011

Citing 9
Cited 1

Selective fusion of heterogeneous classifiers

Intelligent Data Analysis
Adaptive mixtures of local experts

Neural Computation
A Comparison of Hybrid ARMA-Elman Models with Single Models for Forecasting Interest Rates

IITA '08 Proceedings of the 2008 Second International Symposium on Intelligent Information Technology Application - Volume 02
Feature subset selection in large dimensionality domains

Pattern Recognition
Generalized regression neural network in modelling river sediment yield

Advances in Engineering Software
Predicting stock returns by classifier ensembles

Applied Soft Computing
Radial basis function neural networks to foresee aftershocks in seismic sequences related to large earthquakes

ICONIP'06 Proceedings of the 13th international conference on Neural Information Processing - Volume Part II
Stock time series forecasting using support vector machines employing analyst recommendations

ISNN'06 Proceedings of the Third international conference on Advances in Neural Networks - Volume Part III
A general regression neural network

IEEE Transactions on Neural Networks

Time-dependent series variance learning with recurrent mixture density networks

Neurocomputing

Quantified Score

Hi-index	0.01

Visualization

Abstract

We propose a novel homogeneous neural network ensemble approach called Generalized Regression Neural Network (GEFTS-GRNN) Ensemble for Forecasting Time Series, which is a concatenation of existing machine learning algorithms. GEFTS uses a dynamic nonlinear weighting system wherein the outputs from several base-level GRNNs are combined using a combiner GRNN to produce the final output. We compare GEFTS with the 11 most used algorithms on 30 real datasets. The proposed algorithm appears to be more powerful than existing ones. Unlike conventional algorithms, GEFTS is effective in forecasting time series with seasonal patterns.