Median-Pi artificial neural network for forecasting
- 207 Downloads
Datasets with outliers can be predicted with robust learning methods or robust artificial neural networks. In robust artificial neural networks, the architectures become robust by using robust statistics as aggregation functions. Median neural network and trimmed mean neural network are two robust artificial neural networks used in the literature. In these robust artificial neural networks, median and trimmed mean statistics are used as aggregation functions. In this study, Median-Pi artificial neural network is proposed as a new robust neural network for the purpose of forecasting. In Median-Pi artificial neural network, median and multiplicative functions are used as aggregation functions. Because of using median, the proposed network can produce good results for data with outliers. The Median-Pi artificial neural network is trained by particle swarm optimization. The performance of the neural network is investigated by using datasets from the International Time Series Forecast Competition 2016 (CIF-2016). The performance of the proposed method in case of outlier is compared to some other artificial neural networks. Median neural network, trimmed mean neural network, Pi-Sigma neural network and the proposed robust network are applied to time series with outlier, and the obtained results are compared. According to application results, the proposed Median-Pi artificial neural network can produce better forecast results than the other network types.
KeywordsForecasting Outliers Particle swarm optimization Robust artificial neural networks
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
- 2.Hsiao C-C, Chuang C-C, Jeng J-T (2012) Robust back propagation learning algorithm based on near sets. In: 2012 Int. Conf. Syst. Sci. Eng. (ICSSE),June 30–July 2, 2012, Dalian, China. pp 19–23Google Scholar
- 4.El-Melegy MT, Essai MH, Ali AA (2009) Robust training of artificial feedforward neural networks. In: Found. Comput. Intell. Vol. 1 Vol. 201 Ser. Stud. Comput. Intell. pp 217–242Google Scholar
- 15.Attia MA, Sallam EA, Fahmy MM. (2012) A proposed generalized mean single multiplicative neuron model. Proceedings—2012 I.E. 8th International Conference on Intelligent Computer Communication and Processing, ICCP 2012, art. no. 6356163, pp. 73–78Google Scholar
- 27.Kennedy J, Eberhart R, Coello CAC, et al (1995) Particle swarm optimization. In: Neural Networks, 1995. Proceedings., IEEE Int. Conf. pp 1942–1948 vol.4Google Scholar
- 28.Shi Y, Eberhart RC (1999) Empirical study of particle swarm optimization. Evol Comput 1999 CEC 99 Proc 1999 Congr 3:1–1950 Vol. 3. doi: 10.1109/CEC.1999.785511
- 29.Yuchao M, Chuanwen J, Zhijian H, Chenming W (2006) The formulation of the optimal strategies for the electricity producers based on the particle swarm optimization algorithm. Power Syst IEEE Trans 21:1663–1671Google Scholar