A comparison of artificial neural network and time series models for forecasting commodity prices

Abstract A feedforward neural network which can account for nonlinear relationships was used to compare ARIMA and neural network price forecasting performance. Data used was monthly live cattle and wheat prices from 1950 through 1990. The experiment was repeated seven times for successive three year periods. This involved using a walk forward or sliding window approach from 1970 through 1990 which generated out of sample results. The neural network models achieved a 27 percent and 56 percent lower mean squared error than ARIMA model. The absolute mean error and mean absolute percent error were also lower for the neural network models. The neural network models were able to capture a significant number of turning points for both wheat and cattle, while the ARIMA model was only able to do so for wheat. Since this forecasting method is not problem specific and uses only past prices, it can be applied to other forecasting problems such as stocks and other financial prices.

[1]  B. LeBaron,et al.  Nonlinear Dynamics, Chaos, and Instability: Statistical Theory and Economic Evidence , 1991 .

[2]  B. Wade Brorsen,et al.  Nonlinear Dynamics of Daily Cash Prices , 1992 .

[3]  R. Cumby,et al.  Testing for market timing ability: A framework for forecast evaluation , 1987 .

[4]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1992, Math. Control. Signals Syst..

[5]  Halbert White,et al.  On learning the derivatives of an unknown mapping with multilayer feedforward networks , 1992, Neural Networks.

[6]  Christopher S. McIntosh,et al.  Qualitative Forecast Evaluation: A Comparison of Two Performance Measures , 1992 .

[7]  B. LeBaron,et al.  Nonlinear Dynamics and Stock Returns , 2021, Cycles and Chaos in Economic Equilibrium.

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

[9]  Ken'ichi Kamijo,et al.  Stock price pattern recognition-a recurrent neural network approach , 1990, 1990 IJCNN International Joint Conference on Neural Networks.

[10]  Melody Y. Kiang,et al.  Managerial Applications of Neural Networks: The Case of Bank Failure Predictions , 1992 .

[11]  Eberhard Schöneburg,et al.  Stock price prediction using neural networks : A project report , 2003 .

[12]  R. Hecht-Nielsen,et al.  Neurocomputing: picking the human brain , 1988, IEEE Spectrum.

[13]  Jean-Paul Chavas,et al.  Market Instability and Nonlinear Dynamics , 1993 .

[14]  A. Lapedes,et al.  Nonlinear Signal Processing Using Neural Networks , 1987 .

[15]  B. Wade Brorsen,et al.  Nonlinear dynamics of daily futures prices: Conditional heteroskedasticity or chaos? , 1993 .

[16]  Steven c. Blank.  "Chaos" In Futures Markets? A Nonlinear Dynamical Analysis , 1991 .

[17]  Raymond M. Leuthold,et al.  A Comparison of Alternative Forecasting Techniques for Livestock Prices: A Case Study , 1985 .

[18]  Christopher S. McIntosh,et al.  Results of a Price Forecasting Competition , 1990 .

[19]  Kishan G. Mehrotra,et al.  Forecasting the behavior of multivariate time series using neural networks , 1992, Neural Networks.

[20]  James L. McClelland,et al.  Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[21]  William A. Barnett,et al.  Dynamic econometric modeling: The aggregation-theoretic monetary aggregates are chaotic and have strange attractors: an econometric application of mathematical chaos , 1988 .

[22]  Farmer,et al.  Predicting chaotic time series. , 1987, Physical review letters.

[23]  James L. McClelland,et al.  Explorations in parallel distributed processing: a handbook of models, programs, and exercises , 1988 .

[24]  David A. Bessler,et al.  Forecasting livestock prices with individual and composite methods , 1981 .

[25]  Duane DeSieno,et al.  Trading Equity Index Futures With a Neural Network , 1992 .

[26]  David Hsieh.  Chaos and Nonlinear Dynamics: Application to Financial Markets , 1991 .

[27]  Alvin J. Surkan,et al.  Neural networks for bond rating improved by multiple hidden layers , 1990, 1990 IJCNN International Joint Conference on Neural Networks.

[28]  R. Palmer,et al.  Introduction to the theory of neural computation , 1994, The advanced book program.

[29]  T. Willey.  Testing for nonlinear dependence in daily stock indices , 1992 .

[30]  Andrew Harvey,et al.  The econometric analysis of time series , 1991 .

[31]  George E. P. Box,et al.  Time Series Analysis: Forecasting and Control , 1977 .

[32]  Ramesh Sharda,et al.  A neural network model for bankruptcy prediction , 1990, 1990 IJCNN International Joint Conference on Neural Networks.