An Investigation of Differencing Effect in Multiplicative Neuron Model Artificial Neural Network for Istanbul Stock Exchange Time Series Forecasting

In recent years, good alternative methods have been proposed to obtain forecasts for a time series. Artificial neural networks have been commonly used for forecasting purpose in the literature. Although, multilayer perceptron artificial neural network is the most used artificial neural network type, multiplicative neuron model artificial neural networks have been used to obtain forecasts for six years. In the literature, many studies used original series without applying any differencing operation. Thereby, non-stationary time series were used in the artificial neural networks. It is very difficult to find appropriate time series model for non-stationary time series in probabilistic time series methods. Similarly, differencing can be useful obtaining forecasts by artificial neural networks. Differencing effect has not been sufficiently discussed for artificial neural networks. It has not been discussed for multiplicative neuron model artificial network in the literature, yet. Aim of this study is discussing of differencing effect for multiplicative neuron model artificial neural networks. Istanbul stock exchange data (IEX) sets were used to explore differencing effect. The data setsare made up of five time series for years between 2009 and 2013. All of time series were daily observed and observations of them are taken for first five months. It is shown that differencing operation is not useful for forecasting IEX as a result of statistical hypothesis tests.

[1]  Erol Egrioglu,et al.  A Novel Seasonal Fuzzy Time Series Method to the Forecasting of Air Pollution Data in Ankara , 2013 .

[2]  Carlos E. Pedreira,et al.  Neural networks for short-term load forecasting: a review and evaluation , 2001 .

[3]  Erol Egrioglu,et al.  Comparisons of Logistic Regression and Artificial Neural Networks in Lung Cancer Data , 2013 .

[4]  Riccardo Poli,et al.  Particle swarm optimization , 1995, Swarm Intelligence.

[5]  Erol Egrioglu,et al.  Improvement in Forecasting Accuracy Using the Hybrid Model of ARFIMA and Feed Forward Neural Network , 2012 .

[6]  Liang Zhao,et al.  PSO-based single multiplicative neuron model for time series prediction , 2009, Expert Syst. Appl..

[7]  Çagdas Hakan Aladag,et al.  A New Multiplicative Seasonal Neural Network Model Based on Particle Swarm Optimization , 2012, Neural Processing Letters.

[8]  Prem Kumar Kalra,et al.  Time series prediction with single multiplicative neuron model , 2007, Appl. Soft Comput..

[9]  Dervis Karaboga,et al.  AN IDEA BASED ON HONEY BEE SWARM FOR NUMERICAL OPTIMIZATION , 2005 .

[10]  Abir Jaafar Hussain,et al.  Dynamic Ridge Polynomial Neural Network: Forecasting the univariate non-stationary and stationary trading signals , 2011, Expert Syst. Appl..

[11]  Tae Yoon Kim,et al.  Artificial neural networks for non-stationary time series , 2004, Neurocomputing.

[12]  Michael Y. Hu,et al.  Forecasting with artificial neural networks: The state of the art , 1997 .

[13]  Erol Eğrioğlu,et al.  COMPARISON OF SINGLE MULTIPLICATIVE NEURON ARTIFICIAL NEURAL NETWORK MODELS USING ABC AND BP TRAINING ALGORITHMS , 2013 .

[14]  Masataka Sawada,et al.  A Hybrid Measurement System of Three Dimensional Coordinates by Combination of a Multi-link Manipulator and Particle Swarm Optimization Techniques , 2013 .

[15]  Çagdas Hakan Aladag,et al.  A new linear & nonlinear artificial neural network model for time series forecasting , 2013, Decis. Support Syst..

[16]  Tommy W. S. Chow,et al.  Neural network based short-term load forecasting using weather compensation , 1996 .

[17]  Dervis Karaboga,et al.  Artificial Bee Colony (ABC) Optimization Algorithm for Training Feed-Forward Neural Networks , 2007, MDAI.

[18]  B. Samanta,et al.  Prediction of chaotic time series using computational intelligence , 2011, Expert Syst. Appl..

[19]  Karin Kandananond,et al.  Applying 2k Factorial Design to Assess the Performance of ANN and SVM Methods for Forecasting Stationary and Non-stationary Time Series , 2013, KES.

[20]  Erol Egrioglu,et al.  An Statistical Research on Feed Forward Neural Networks for Forecasting Time Series , 2012 .