Feedforward and Recurrent Neural Networks for Time Series Forecasting: Comparative Study

This study aims at examining and comparing the ability of ANNs variances, including MLP, RBFNN, ELMAN and JORDAN in forecasting monthly data of four different time series patterns. As well as to criticize the concept asserting that MLP is a "universal approximator". The tackled patterns include time series with seasonality, trend, combined seasonality and trend, and non-constant variance. Thereafter, based on statistical metrics, the results of this study showed that ELMAN network outperforms the remaining models in dealing with all types of time series patterns. Therefore, we can confirm that ELMAN network is more suited to time series forecasting than feedforward networks, thanks to its embedding memory between its input and hidden layers.

[1]  Malek Sarhani,et al.  Hybrid approach-based support vector machine for electric load forecasting incorporating feature selection , 2017, Int. J. Big Data Intell..

[2]  Malek Sarhani,et al.  Forecasting Demand With Support Vector Regression Technique Incorporating Feature Selection in the Presence of Calendar Effect , 2018 .

[3]  Jae Kyu Lee,et al.  Performance of Neural Networks in Managerial Forecasting , 1993 .

[4]  Guoqiang Peter Zhang,et al.  Neural network forecasting for seasonal and trend time series , 2005, Eur. J. Oper. Res..

[5]  K. Chau,et al.  Improving forecasting accuracy of medium and long-term runoff using artificial neural network based on EEMD decomposition. , 2015, Environmental research.

[6]  Malek Sarhani,et al.  Electric Load Forecasting Using Hybrid Machine Learning Approach Incorporating Feature Selection , 2015, BDCA.

[7]  Sheng Chen,et al.  Recursive hybrid algorithm for non-linear system identification using radial basis function networks , 1992 .

[8]  J. J. Montaño Moreno,et al.  Artificial neural networks applied to forecasting time series. , 2011, Psicothema.

[9]  J. Moreno,et al.  Artificial neural networks applied to forecasting time series , 2011 .

[10]  Youngdeok Hwang,et al.  Artificial neural network model for forecasting sub-hourly electricity usage in commercial buildings , 2016 .

[11]  Guoqiang Peter Zhang,et al.  Trend time series modeling and forecasting with neural networks , 2003, 2003 IEEE International Conference on Computational Intelligence for Financial Engineering, 2003. Proceedings..

[12]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[13]  Abdellatif El Afia,et al.  Prediction of Supplier Performance: A Novel Dea-Anfis Based Approach , 2017, BDCA'17.

[14]  Abdellatif El Afia,et al.  Performance prediction of pharmaceutical suppliers: comparative study between DEA-ANFIS-PSO and DEA-ANFIS-GA , 2019, Int. J. Comput. Appl. Technol..

[15]  João Paulo Teixeira,et al.  Tourism Time Series Forecast -Different ANN Architectures with Time Index Input , 2012, Procedia Technology.

[16]  Amir F. Atiya,et al.  An Empirical Comparison of Machine Learning Models for Time Series Forecasting , 2010 .

[17]  Werner Kristjanpoller,et al.  Volatility forecast using hybrid Neural Network models , 2014, Expert Syst. Appl..

[18]  Malek Sarhani,et al.  Intelligent system based support vector regression for supply chain demand forecasting , 2014, 2014 Second World Conference on Complex Systems (WCCS).

[19]  Malek Sarhani,et al.  Feature selection and parameter optimization of support vector regression for electric load forecasting , 2016, 2016 International Conference on Electrical and Information Technologies (ICEIT).

[20]  Rob J Hyndman,et al.  Forecasting with Exponential Smoothing: The State Space Approach , 2008 .

[21]  Malek Sarhani,et al.  Forecasting Demand with Support Vector Regression Technique Combined with X13-ARIMA-SEATS Method in the Presence of Calendar Effect , 2014, Int. J. Appl. Logist..

[22]  Abdellatif El Afia,et al.  Artificial Neural Network Based Approach for Blood Demand Forecasting: Fez Transfusion Blood Center Case Study , 2017, BDCA'17.

[23]  J. A. Leonard,et al.  Radial basis function networks for classifying process faults , 1991, IEEE Control Systems.