A novel neural network ensemble architecture for time series forecasting
NeurocomputingPublished 29 August 2011
Iffat A. Gheyas, Leslie S. Smith
Citations73
SJR quartileQ1
SJR score1.47
SNIP1.94
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
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.
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 use 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.
Keywords
Computer ScienceDecision SciencesEngineering
Revue économiqueNonparametric Statistics for the Behavioral Sciences
35,492 Citations1958Henri Guitton, Sidney Siegel
Neural ComputationAdaptive Mixtures of Local Experts
4,988 Citations1991Robert A. Jacobs, Michael I. Jordan +2 more
A new supervised learning procedure for systems composed of many separate networks, each of which learns to handle a subset of the complete set of training cases, which is demonstrated to be able to be solved by a very simple expert network.
IEEE Transactions on Neural NetworksA general regression neural network
4,415 Citations1991Donald F. Specht
The general regression neural network (GRNN) is a one-pass learning algorithm with a highly parallel structure that provides smooth transitions from one observed value to another.
Pattern RecognitionFeature subset selection in large dimensionality domains
592 Citations2009Iffat A. Gheyas, Leslie S. Smith
This work presents a hybrid algorithm, SAGA, that combines the ability to avoid being trapped in a local minimum of simulated annealing with the very high rate of convergence of the crossover operator of genetic algorithms, the strong local search ability of greedy algorithms and the high computational efficiency of generalized regression neural networks.
Advances in Engineering SoftwareGeneralized regression neural network in modelling river sediment yield
332 Citations2005H. Kerem Ciğizoğlu, Murat Alp
Another ANN algorithm, generalized regression neural network, GRNN, was used in river suspended sediment estimation, and the ANN estimations are found significantly superior to conventional method results.
Water Resources ManagementRiver Flow Forecasting using Recurrent Neural Networks
251 Citations2004D. Nagesh Kumar, K. Srinivasa Raju +1 more
Recurrent neural networks were used to train and forecast the monthly flows of a river in India, with a catchment area of 5189 km2 up to the gauging site, and performed better than the feed forward networks.
Applied Soft ComputingPredicting stock returns by classifier ensembles
169 Citations2010Chih‐Fong Tsai, Yuah-Chiao G. Lin +2 more
This study aims at investigating the prediction performance that utilizes the classifier ensembles method to analyze stock returns and indicates that multiple classifiers outperform single classifiers in terms of prediction accuracy and returns on investment.
Intelligent Data AnalysisSelective fusion of heterogeneous classifiers
94 Citations2005Grigorios Tsoumakas, Lefteris Angelis +1 more
The proposed approach, Selective Fusion, improves over simple selection and fusion methods, leading to performance comparable with the state-of-the-art heterogeneous classifier combination method of Stacking, without the additional computational cost and learning problems of meta-training.
Lecture notes in computer scienceTime Series Prediction with Multilayer Perceptron (MLP): A New Generalized Error Based Approach
44 Citations2009Mohammad Shiblee, Prem Kalra +1 more
Wide variety of well known time series prediction problems that generalized geometric and harmonic error measures perform better than LMSE for wide class of problems are illustrated.
International Journal of ForecastingPrediction intervals in conditionally heteroscedastic time series with stochastic components
25 Citations2010Santiago Pellegrini, Esther Ruiz +1 more
The Journal of Economic EducationEconomic Time-Series Page
17 Citations2000Theodore Bos, Sarah E. Culver
Lecture notes in computer scienceOptimizing Weights by Genetic Algorithm for Neural Network Ensemble
16 Citations2004Zhangquan Shen, Fansheng Kong
This paper presents an ensemble method for regression that has advantages over simple weighted or weighted average combining techniques and achieved high predicting accuracy on five test datasets.
Lecture notes in computer scienceStock Time Series Forecasting Using Support Vector Machines Employing Analyst Recommendations
13 Citations2006Zhiyong Zhang, Chuan Shi +2 more
An improved method to evaluate if an upgrade classification of SVM is reliable is put forward, and some interesting hints are discussed to help readers understand this model more explicitly.
Lecture notes in computer scienceRadial Basis Function Neural Networks to Foresee Aftershocks in Seismic Sequences Related to Large Earthquakes
9 Citations2006Vincenzo Barrile, Matteo Cacciola +4 more
Radial Basis Function Neural Network was implemented in order to predict the trend of n(t) for aftershocks temporal series, that is the numerical series of daily-earthquake's number occurred after a great earthquake with magnitude M > 7.0 Richter.
A Comparison of Hybrid ARMA-Elman Models with Single Models for Forecasting Interest Rates
5 Citations2008Xiaojian Yu, Jiaping Zhang
Using the data of benchmark interest rates of China, the empirical results indicate that the hybrid models are superior to the single ARMA model, but perform closely to the Elman recurrent neural network model.