Methodological Advances in Artificial Neural Networks for Time Series Forecasting
IEEE Latin America TransactionsPublished 1 June 2014Open access
Myladis R. Cogollo, Juan D. Velásquez
Citations16
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
It is found that the studies proposing new forecasting models based on neural networks with a theoretical support and a systematic procedure for the construction of model, were scarce in the time period 2000-2010.
Abstract
Objective: The aim of this paper is to analyze the development of new forecasting models based on neural networks. Method: We used the systematic literature review method employing a manual search of papers published on new neural networks models in the time period 2000 to 2010. Results: Only 18 studies meet all the requirements of the inclusion criteria. Of these, only three proposals considered a neural networks model using a process different to the autoregressive. Conclusion: Although studies relating to the application of neural network models were frequently present, we find that the studies proposing new forecasting models based on neural networks with a theoretical support and a systematic procedure for the construction of model, were scarce in the time period 2000-2010.
Keywords
Decision Sciences
EconometricaAutoregressive Conditional Heteroscedasticity with Estimates of the Variance of United Kingdom Inflation
20,637 Citations1982Robert F. Engle
NeurocomputingTime series forecasting using a hybrid ARIMA and neural network model
4,350 Citations2003G.Peter Zhang
Experimental results with real data sets indicate that the combined model can be an effective way to improve forecasting accuracy achieved by either of the models used separately.
Practical neural network recipes in C
1,653 Citations1993Timothy Masters
International Journal of Forecasting25 years of time series forecasting
1,565 Citations2006Jan G. De Gooijer, Rob J. Hyndman
Cambridge University Press eBooksNonlinear Time Series Models in Empirical Finance
1,130 Citations2000Philip Hans Franses, Dick van Dijk
NeurocomputingDesigning a neural network for forecasting financial and economic time series
1,091 Citations1996Iebeling Kaastra, Milton S. Boyd
An eight-step procedure to design a neural network forecasting model is explained including a discussion of tradeoffs in parameter selection, some common pitfalls, and points of disagreement among practitioners.
Expert Systems with ApplicationsAn artificial neural network (p,d,q) model for timeseries forecasting
785 Citations2009Mehdi Khashei, Mehdi Bijari
The empirical results with three well-known real data sets indicate that the proposed model can be an effective way to improve forecasting accuracy achieved by artificial neural networks, and can be used as an appropriate alternative model for forecasting task, especially when higher forecasting accuracy is needed.
Expert Systems with ApplicationsNeural networks and statistical techniques: A review of applications
773 Citations2007Mukta Paliwal, Usha A. Kumar
A comprehensive review of articles that involve a comparative study of feed forward neural networks and statistical techniques used for prediction and classification problems in various areas of applications is carried out.
Journal of EconometricsTesting for neglected nonlinearity in time series models
507 Citations1993Tae‐Hwy Lee, Halbert White +1 more
The results suggest that the neural network test can play a valuable role in evaluating model adequacy, and has proper size and good power, and many of the economic series tested exhibit potential nonlinearities.
The MIT Press eBooksFoundations of Neural Networks, Fuzzy Systems, and Knowledge Engineering
478 Citations1996Nikola Kasabov
Journal of Time Series AnalysisPOWER OF THE NEURAL NETWORK LINEARITY TEST
377 Citations1993Timo Teräsvirta, Chien‐Fu Lin +1 more
The power properties of a new linearity test for time series based on concepts from the theory of neural networks are investigated by simulation with those of a Lagrange multiplier (LM) type test that derives from the same single‐hidden‐layer neural network model.
Expert Systems with ApplicationsA fusion model of HMM, ANN and GA for stock market forecasting
358 Citations2006Md. Rafiul Hassan, Baikunth Nath +1 more
A fusion model by combining the Hidden Markov Model (HMM), Artificial Neural Networks (ANN) and Genetic Algorithms (GA) to forecast financial market behaviour is proposed and implemented.
Neural NetworksModel selection in neural networks
355 Citations1999Ulrich Anders, Olaf Korn
This article examines how model selection in neural networks can be guided by statistical procedures such as hypothesis tests, information criteria and cross validation, and proposes five specification strategies based on different statistical procedures.
Technological Forecasting and Social ChangeCombining neural network model with seasonal time series ARIMA model
342 Citations2002Fang-Mei Tseng, Hsiao–Cheng Yu +1 more
A hybrid forecasting model, which combines the seasonal time series ARIMA (SARIMA) and the neural network back propagation (BP) models, known as SARIMABP, was used to forecast two seasonal timeseries data of total production value for Taiwan machinery industry and the soft drink time series.
Information SciencesTime-series forecasting using flexible neural tree model
294 Citations2004Yuehui Chen, Bo Yang +2 more
This paper introduces a new time-series forecasting model based on the flexible neural tree (FNT), which is generated initially as a flexible multi-layer feed-forward neural network and evolved using an evolutionary procedure.
International Journal of ForecastingForecasting economic and financial time-series with non-linear models
292 Citations2004Michael P. Clements, Philip Hans Franses +1 more
This paper discusses the current state-of-the-art in estimating, evaluating, and selecting among non-linear forecasting models for economic and financial time series, and discusses a number of areas which have received considerable attention in the recent literature, but where many questions remain.
Computers & Operations ResearchA novel nonlinear ensemble forecasting model incorporating GLAR and ANN for foreign exchange rates
260 Citations2004Lean Yu, Shouyang Wang +1 more
The findings reveal that the nonlinear ensemble model proposed here can be used as an alternative forecasting tool for exchange rates to achieve greater forecasting accuracy and improve prediction quality further.
European Journal of Operational ResearchAn investigation of model selection criteria for neural network time series forecasting
253 Citations2001Min Qi, Guoqiang Peter Zhang
Results indicate that the in-sample model selection criteria investigated are not able to provide a reliable guide to out-of-sample performance and there is no apparent connection between in- sample model fit and out- of-sample forecasting performance.
Computers & StructuresPrediction of principal ground-motion parameters using a hybrid method coupling artificial neural networks and simulated annealing
241 Citations2011Amir H. Alavi, Amir H. Gandomi
An additional hidden unit test for neglected nonlinearity in multilayer feedforward networks
210 Citations1989White
The author presents a statistical test of the hypothesis that a given multilayer feedforward network exactly represents some unknown mapping subject to inherent noise against the alternative that the network neglects some nonlinear structure in the mapping, leading to potentially avoidable approximation errors.
Fuzzy Sets and SystemsA new hybrid artificial neural networks and fuzzy regression model for time series forecasting
204 Citations2007Mehdi Khashei, Seyed Reza Hejazi +1 more
Based on the basic concepts of ANNs and fuzzy regression models, a new hybrid method is proposed that yields more accurate results with incomplete data sets and the empirical results of financial market forecasting indicate that the proposed model can be an effective way of improving forecasting accuracy.
RePEc: Research Papers in EconomicsBuilding neural network models for time series: a statistical approach
185 CitationsMarcelo C. Medeiros, Timo Teräsvirta +1 more
Journal of HydrologyEvolutionary artificial neural networks for hydrological systems forecasting
173 Citations2009Yung‐Hsiang Chen, Fi‐John Chang
A novel evolutionary artificial neural network (EANN) for time series forecasting that has a hybrid procedure, including the genetic algorithm and the scaled conjugate gradient algorithm, where the feedforward ANN architecture and its connection weights of neurons are simultaneously identified and optimized.
IEEE Transactions on Power SystemsA Hybrid Model for Day-Ahead Price Forecasting
171 Citations2010Lei Wu, Mohammad Shahidehpour
This paper presents a hybrid time-series and adaptive wavelet neural network (AWNN) model for the day-ahead electricity market clearing price forecast that uses one-period continuously compounded return series instead of using price series to achieve more attractive statistical properties.
Systems Research and Behavioral ScienceAn ARIMA‐ANN Hybrid Model for Time Series Forecasting
167 Citations2013Li Wang, Zou Haofei +3 more
A hybrid model is proposed, which is distinctive in integrating the advantages of ARIMA and ANNs in modeling the linear and nonlinear behaviors in the data set, and indicates the effectiveness of the new combinatorial model in obtaining more accurate forecasting as compared to existing models.
European Journal of Operational ResearchAdaptive neural network model for time-series forecasting
104 Citations2010Wai Keung Wong, Min Xia +1 more
The proposed ADNN method can predict periodical time-series with a complicated structure and outperforms the auto-regression (AR), artificial neural network (ANN), and adaptive k-nearest neighbors (AKN) models.
NeurocomputingA dynamic architecture for artificial neural networks
103 Citations2004M. Ghiassi, H. Saidane
This paper introduces a model that uses a different architecture compared to the traditional neural network, to capture and forecast nonlinear processes, and shows that this approach performs well when compared with traditional models and established research.
Econstor (Econstor)Forecasting economic variables with nonlinear models
96 Citations2005Timo Teräsvirta
NeurocomputingA novel neural network ensemble architecture for time series forecasting
73 Citations2011Iffat A. Gheyas, Leslie S. Smith
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.
IEEE Transactions on Neural NetworksA hybrid linear-neural model for time series forecasting
71 Citations2000A. Veiga, Marcelo C. Medeiros
A linear model with time varying parameters controlled by a neural network to analyze and forecast nonlinear time series with the advantage of naturally incorporating linear multivariate thresholds and smooth transitions between regimes is considered.
Fire Safety JournalA novel artificial neural network fire model for prediction of thermal interface location in single compartment fire
69 Citations2003Eric Wai Ming Lee, Richard K.K. Yuen +3 more
A unique, new and novel artificial neural network model, denoted as GRNNFA, is developed for predicting parameters in compartment fires and is an extremely fast alternative approach for capturing the nonlinear system behaviour by training the network using relevant historical data.
IEEE Transactions on Neural NetworksA Flexible Coefficient Smooth Transition Time Series Model
65 Citations2005Marcelo C. Medeiros, A. Veiga
If the neural network is interpreted as a nonparametric universal approximation to any Borel measurable function, this formulation is directly comparable to the functional coefficient autoregressive (FAR) and the single-index coefficient regression models.
Chaos An Interdisciplinary Journal of Nonlinear ScienceThe architecture of dynamic reservoir in the echo state network
52 Citations2012Hongyan Cui, Xiang Liu +1 more
Three novel types of echo state networks with new dynamic reservoir topologies based on complex network theory with better prediction capabilities, a wider spectral radius, but retain almost the same short-term memory capacity as compared to the conventional echo state network model.
Journal of the American Statistical AssociationStochastic Neural Networks With Applications to Nonlinear Time Series
38 Citations2001Tze Leung Lai, Samuel Po‐Shing Wong
It is shown how the expectation-maximization algorithm can be used to develop efficient estimation schemes that have much lower computational complexity than those for conventional neural networks.
Neural Computing and ApplicationsApplications of AR*-GRNN model for financial time series forecasting
34 Citations2007Weimin Li, Yishu Luo +3 more
A model composed of AR* and GRNN is proposed to take advantage of their feathers in linear and nonlinear modeling and provides an effective way to improve forecasting performance which can be achieved by either of the models used separately.
Journal of HydrologyA nonlinear perturbation model based on artificial neural network
25 Citations2006Bo Pang, Shenglian Guo +2 more
A nonlinear perturbation model based on artificial neural network (ANN), defined as NLPM-ANN, for the purpose of improving the rainfall–runoff forecasting efficiency and accuracy is developed and it is shown that the model efficiency is significantly higher than that of the LPM.
Signal ProcessingModelling non-linear moving average processes using neural networks with error feedback: An application to implied volatility forecasting
17 Citations1999Andrew N. Burgess, Apostolos P. Refenes
It is demonstrated that although in principle a (sufficiently high- order) auto-regressive process can approximate the MA process, in practice the diminishing information content of the high-order AR terms will at some point be offset by the increase in model variance caused by the additional model parameters, leading to a degradation in generalisation performance.