Advertisement

Forecasting Cash Demand in ATM Using Neural Networks and Least Square Support Vector Machine

  • Cristián Ramírez
  • Gonzalo Acuña
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7042)

Abstract

In this work we forecast the daily ATM cash demand using dynamic models of type Nonlinear Autoregressive Exogeneous inputs (NARX) and Nonlinear Autoreggressive Moving Average with Exogeneous Inputs (NARMAX) performed by Neural Networks (NN) and Least Square Support Vector Machine (LS-SVM) and used to predict one step (OSA) or multistep (MPO). The aim is to compare which model perform better results. We found that the Multilayer Perceptron NN presented the best index of agreement with an average of 0.87 in NARX-OSA and 0.85 in NARX-MPO. After, Radial Basis Function NN was 0.82 for both cases. Finally, LS-SVM obtained the worst results with 0.78 for NARX-OSA and 0.70 for NARX-MPO. No significant differences between NARX and NARMAX structures were found. Our contribution would have obtained the 2 nd place in the NN5 competition of computational methods.

Keywords

MLP RBF LS-SVM NARX NARMAX OSA MPO NN5 IA SMAPE 

References

  1. 1.
    Simutis, R., Dilijonas, D., Bastina, L.: Cash demand forecasting for ATM using neural Networks and support vector regresin algorithms. In: 20th EURO Mini Conference, Vilnius, pp. 416–421 (2008)Google Scholar
  2. 2.
    Simutis, R., Dilijonas, D., Bastina, L., Friman, J., Drobinov, P.: Optimization of Cash Management for ATM Network. Information Technology and Control 36(1A) (2007) ISSN 1392 124x Google Scholar
  3. 3.
    Justine, P.: ATMs and Cash Demand Forecasting: A Study of Two Commercial Bank. Nagoya University of Commerce and Business, 1-20-1, Nishiki, Naka-Ku, Nagoya City, JAPAN- 460 0003 (2007)Google Scholar
  4. 4.
    Velasquez, E.R.: Redes neuronales para sistemas dinmicos tipo NOE y Narmax. Tesis de magster, Universidad de Santiago de Chile. Departamento de Informtica (2007)Google Scholar
  5. 5.
    Ljung, L.: System Identification: Theory for the User. Prentice Hall, Englewood Cliffs (1987)zbMATHGoogle Scholar
  6. 6.
    NN5, Forecasting Competition for Artificial Neural Networks & Computational Intelligence (2008), http://www.neural-forecastingcompetition.com/NN5
  7. 7.
    Jorg, D., Wichard: Forecasting the NN5 time series with hybrid models. Leibniz-Institut für Molekulare Pharmakologie (FMP), Molecular Modeling Group, Robert-Roessle-Str. 10, 13125, Berlin, Germany (2010)Google Scholar
  8. 8.
    Teddy, S.D.: Forecasting ATM Cash Demand using a local learning model of a cerebellar associative memory network. Data Mining Department, Institute for Infocomm Research Singapore 138632, Singapore (2010)Google Scholar
  9. 9.
    Tukey, J.W.: Exploratory Data Analisis. Addison Wesley (1977)Google Scholar
  10. 10.
    Bouchachia, A., Bouchachia, S.: Ensemble Learning for Time Series Prediction. In: First International Workshop on Nonlinear Dynamics Synchronization, Austria, INDS (2008)Google Scholar
  11. 11.
    Gao, Y., Joo Er, M.: NARMAX time series model prediction: feedfoward and recurrent fuzzy neural network approaches. Fuzzy Sets and Systems 150, 331–350 (2005)MathSciNetCrossRefzbMATHGoogle Scholar
  12. 12.
    Zemouri, R., Racoceanu, D., Zerhouni, N.: Recurrent radial basis function network for time-series prediction. In: Engineering Applications of Artificial Intelligence, pp. 453–463 (2003)Google Scholar
  13. 13.
    Espinoza, M., Falck, T., Suykens, J., De Moor, B.: Time Series Prediction using LS-SVMs. Kasteelpark Arenberg 10, B-3001 Leuven, Heverlee, Belgium (2008)Google Scholar
  14. 14.
    Garcia-Pedrero, A., Gomez-Gil, P.: Time Series Forecasting using Recurrent Neural Networks and Wavelet Reconstructed Signals. In: 20th International Conference on Electronics, Communications and Computer (CONIELECOMP) (2010)Google Scholar
  15. 15.
    Rode, M., Suhr, U.: Multi-Objective Calibration of a River Water Quality Model For The Elbe River, Germany, UFZ Centre for Environmental Research Leipzig Halle, Department of Hydrological Modelling, Brckstrae 3a, D-39114 Magdeburg, GermanyGoogle Scholar
  16. 16.
    Norgaard, M.: Neural Networks for Modelling and Control of Dynamic Systems: A Practitioner’s Handbook (Advanced Textbooks in Control and Signal Processing). Springer, Heidelberg (2006)Google Scholar
  17. 17.
    Soares, P., Rodriguez, P.: Calendar Effects in Daily ATM Withdrawals. Economics and Research Department. Banco de Portugal (2010)Google Scholar
  18. 18.
    Chen, P.-H., Lee, Y.-W., Chang, T.-L.: Predicting thermal instability in a closed loop pulsating heat pipe system. Applied Thermal Engineering (2008)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Cristián Ramírez
    • 1
  • Gonzalo Acuña
    • 1
  1. 1.Departamento de Ingeniería InformáticaUniversidad de Santiago de Chile, USACHOhigginsChile

Personalised recommendations