Home > Research > Publications & Outputs > Input-variable Specification for Neural Network...
View graph of relations

Input-variable Specification for Neural Networks - an Analysis of Forecasting low and high Time Series Frequency

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Published

Standard

Input-variable Specification for Neural Networks - an Analysis of Forecasting low and high Time Series Frequency. / Crone, Sven F.; Kourentzes, Nikolaos.

International Joint Conference on Neural Networks, 2009. IJCNN 2009.. New York : IEEE, 2009. p. 3221-3228.

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Harvard

Crone, SF & Kourentzes, N 2009, Input-variable Specification for Neural Networks - an Analysis of Forecasting low and high Time Series Frequency. in International Joint Conference on Neural Networks, 2009. IJCNN 2009.. IEEE, New York, pp. 3221-3228, International Joint Conference on Neural Networks, Atlanta, 14/06/09. https://doi.org/10.1109/IJCNN.2009.5179046

APA

Vancouver

Crone SF, Kourentzes N. Input-variable Specification for Neural Networks - an Analysis of Forecasting low and high Time Series Frequency. In International Joint Conference on Neural Networks, 2009. IJCNN 2009.. New York: IEEE. 2009. p. 3221-3228 doi: 10.1109/IJCNN.2009.5179046

Author

Crone, Sven F. ; Kourentzes, Nikolaos. / Input-variable Specification for Neural Networks - an Analysis of Forecasting low and high Time Series Frequency. International Joint Conference on Neural Networks, 2009. IJCNN 2009.. New York : IEEE, 2009. pp. 3221-3228

Bibtex

@inproceedings{75d4c0273dcd4346bb4e32b1968bdf76,
title = "Input-variable Specification for Neural Networks - an Analysis of Forecasting low and high Time Series Frequency",
abstract = "Prior research in forecasting time series with Neural Networks (NN) has provided inconsistent evidence on their predictive accuracy. In management, NN have shown only inferior performance on well established benchmark time series of monthly, quarterly or annual frequency. In contrast, NN have shown preeminent accuracy in electrical load forecasting on daily or hourly time series, leading to successful real life applications. While this inconsistency has been traditionally attributed to the lack of a reliable methodology to model NNs, recent research indicates that the particular data properties of high frequency time series may be equally important. High frequency time series of daily, hourly or even shorter time intervals pose additional modelling challenges in the length and structure of the time series, which may abet the use of novel methods. This analysis aims to identify and contrast the challenges in modelling NN for low and high frequency data in order to develop a unifying forecasting methodology tailored to the properties of the dataset. We conduct a set of experiments in three different frequency domains of daily, weekly and monthly data of one empirical time series of cash machine withdrawals, using a consistent modelling procedure. While our analysis provides evidence that NN are suitable to predict high frequency data, it also identifies a set of challenges in modelling NN that arise from high frequency data, in particular in specifying the input vector, and that require specific modelling approaches applicable to both low and high frequency data.",
author = "Crone, {Sven F.} and Nikolaos Kourentzes",
year = "2009",
doi = "10.1109/IJCNN.2009.5179046",
language = "English",
isbn = "978-1-4244-3549-4",
pages = "3221--3228",
booktitle = "International Joint Conference on Neural Networks, 2009. IJCNN 2009.",
publisher = "IEEE",
note = "International Joint Conference on Neural Networks ; Conference date: 14-06-2009 Through 19-06-2009",

}

RIS

TY - GEN

T1 - Input-variable Specification for Neural Networks - an Analysis of Forecasting low and high Time Series Frequency

AU - Crone, Sven F.

AU - Kourentzes, Nikolaos

PY - 2009

Y1 - 2009

N2 - Prior research in forecasting time series with Neural Networks (NN) has provided inconsistent evidence on their predictive accuracy. In management, NN have shown only inferior performance on well established benchmark time series of monthly, quarterly or annual frequency. In contrast, NN have shown preeminent accuracy in electrical load forecasting on daily or hourly time series, leading to successful real life applications. While this inconsistency has been traditionally attributed to the lack of a reliable methodology to model NNs, recent research indicates that the particular data properties of high frequency time series may be equally important. High frequency time series of daily, hourly or even shorter time intervals pose additional modelling challenges in the length and structure of the time series, which may abet the use of novel methods. This analysis aims to identify and contrast the challenges in modelling NN for low and high frequency data in order to develop a unifying forecasting methodology tailored to the properties of the dataset. We conduct a set of experiments in three different frequency domains of daily, weekly and monthly data of one empirical time series of cash machine withdrawals, using a consistent modelling procedure. While our analysis provides evidence that NN are suitable to predict high frequency data, it also identifies a set of challenges in modelling NN that arise from high frequency data, in particular in specifying the input vector, and that require specific modelling approaches applicable to both low and high frequency data.

AB - Prior research in forecasting time series with Neural Networks (NN) has provided inconsistent evidence on their predictive accuracy. In management, NN have shown only inferior performance on well established benchmark time series of monthly, quarterly or annual frequency. In contrast, NN have shown preeminent accuracy in electrical load forecasting on daily or hourly time series, leading to successful real life applications. While this inconsistency has been traditionally attributed to the lack of a reliable methodology to model NNs, recent research indicates that the particular data properties of high frequency time series may be equally important. High frequency time series of daily, hourly or even shorter time intervals pose additional modelling challenges in the length and structure of the time series, which may abet the use of novel methods. This analysis aims to identify and contrast the challenges in modelling NN for low and high frequency data in order to develop a unifying forecasting methodology tailored to the properties of the dataset. We conduct a set of experiments in three different frequency domains of daily, weekly and monthly data of one empirical time series of cash machine withdrawals, using a consistent modelling procedure. While our analysis provides evidence that NN are suitable to predict high frequency data, it also identifies a set of challenges in modelling NN that arise from high frequency data, in particular in specifying the input vector, and that require specific modelling approaches applicable to both low and high frequency data.

U2 - 10.1109/IJCNN.2009.5179046

DO - 10.1109/IJCNN.2009.5179046

M3 - Conference contribution/Paper

SN - 978-1-4244-3549-4

SP - 3221

EP - 3228

BT - International Joint Conference on Neural Networks, 2009. IJCNN 2009.

PB - IEEE

CY - New York

T2 - International Joint Conference on Neural Networks

Y2 - 14 June 2009 through 19 June 2009

ER -