Time-Series Forecasting by Means of Linear and Nonlinear Models

The main objective of this paper is two folds. First is to assess some well-known linear and nonlinear techniques comparatively in modeling and forecasting financial time series with trend and seasonal patterns. Then to investigate the effect of pre-processing procedures, such as seasonal adjustment methods, to the improvement of the modeling capability of a nonlinear structure implemented as ANNs in comparison to the classical Box-Jenkins seasonal autoregressive integrated moving average (ARIMA) model, which is widely used as a linear statistical time series forecasting method. Furthermore, the effectiveness of seasonal adjustment procedures, i.e. direct or indirect adjustments, on the forecasting performance is evaluated. The Autocorrelation Function (ACF) plots are used to determine the correlation between lags due to seasonality, and to determine the number of input nodes that is also confirmed by trial-and-errors. The linear and nonlinear models mentioned above are applied to aggregate retail sales data, which carries strong trend and seasonal patterns. Although, the results without any pre-processing were in an acceptable interval, the overall forecasting performance of ANN was not better than that of the classical method. After employing the right seasonal adjustment procedure, ANN has outperformed its linear counterpart in out-of-sample forecasting. Consequently, it is confirmed that the modeling capability of ANN is improved significantly by using a pre-processing procedure. The results obtained from both ARIMA and ANNs based forecasting methodologies are analyzed and compared with Mann-Whitney statistical test.

[1]  A. Atiya,et al.  Application of neural networks to the problem of forecasting the flow of the River Nile , 1997, Neural Networks for Signal Processing VII. Proceedings of the 1997 IEEE Signal Processing Society Workshop.

[2]  Giorgio Corani,et al.  Air quality prediction in Milan: feed-forward neural networks, pruned neural networks and lazy learning , 2005 .

[3]  A. J. Morris,et al.  Artificial neural network based multivariable predictive control , 1991 .

[4]  David F. Findley,et al.  Comparing Direct and Indirect Seasonal Adjustments of Aggregate Series , 2001 .

[5]  Magali E. Azema-Barac,et al.  Currency exchange rate forecasting by error backpropagation , 1992, Proceedings of the Twenty-Fifth Hawaii International Conference on System Sciences.

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

[7]  Mohammad Bagher Menhaj,et al.  Training feedforward networks with the Marquardt algorithm , 1994, IEEE Trans. Neural Networks.

[8]  Victor R. Prybutok,et al.  Comparison of neural network models with ARIMA and regression models for prediction of Houston's daily maximum ozone concentrations , 2000, Eur. J. Oper. Res..

[9]  Paul A. Fishwick,et al.  Feedforward Neural Nets as Models for Time Series Forecasting , 1993, INFORMS J. Comput..

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

[11]  Philip Hans Franses,et al.  Recognizing changing seasonal patterns using artificial neural networks , 1997 .

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

[13]  Marcus O'Connor,et al.  Artificial neural network models for forecasting and decision making , 1994 .

[14]  Ramesh Sharda,et al.  Connectionist approach to time series prediction: an empirical test , 1992, J. Intell. Manuf..

[15]  Kyungdoo “Ted” Nam,et al.  FORECASTING INTERNATIONAL AIRLINE PASSENGER TRAFFIC USING NEURAL NETWORKS , 1995 .

[16]  William Remus,et al.  Time series forecasting using neural networks: should the data be deseasonalized first? , 1999 .

[17]  M. Qi,et al.  Forecasting Aggregate Retail Sales: a Comparison of Arti"cial Neural Networks and Traditional Methods , 2001 .

[18]  T. Hill,et al.  Neural network models for forecast: a review , 1992, Proceedings of the Twenty-Fifth Hawaii International Conference on System Sciences.

[19]  Seasonal Adjustment Methods : An Application to the Turkish Monetary Aggregates , 2002 .

[20]  Gwilym M. Jenkins,et al.  Time series analysis, forecasting and control , 1971 .

[21]  Suh Young Kang,et al.  An investigation of the use of feedforward neural networks for forecasting , 1992 .

[22]  Jae Kyu Lee,et al.  A neural network approach for the identification of the Box-Jenkins model , 1992 .

[23]  Nejib Smaoui,et al.  The Box-Jenkins analysis and neural networks: prediction and time series modelling , 2003 .

[24]  Yusuf Al-Turki,et al.  A comparative study of medium-weather-dependent load forecasting using enhanced artificial/fuzzy neural network and statistical techniques , 1998, Neurocomputing.