COMPARATIVE ANALYSIS OF THREE TECHNIQUES FOR PREDICTIONS IN TIME SERIES HAVING REPETITIVE PATTERNS

Arash Niknafs, Bo Sun, Michael M. Richter, Günther Ruhe

Abstract

Modelling nonlinear patterns is possible through using regression (curve fitting) methods. However, they can be modelled by linear regression (LR) methods, too. This kind of modelling is usually used to depict and study trends and it is not used for prediction purposes. Our goal is to study the applicability and accuracy of piecewise linear regression in predicting a target variable in different time spans (where a pattern is being repeated). Using moving average, we identified the split points and then tested our approach on a real world case study. The dataset of the amount of recycling material in Blue Carts in Calgary (including more than 31,000 records) was taken as a case study for evaluating the performance of the proposed approach. Root mean square error (RMSE) and Spearman rho were used to evaluate and prove the applicability of this prediction approach and evaluate its performance. A comparison between the performances of Support Vector Machine (SVM), Neural Networks (NN), and the proposed LR-based prediction approach is also presented. The results show that the proposed approach works very well for such prediction purposes. It outperforms SVM and is a powerful competitor for NN.

References

  1. Brown, R. L., Durbin, J. & Evans, J. M. 1975. Techniques for Testing the Constancy of Regression Relationships over Time. Journal of the Royal Statistical Society. Series B (Methodological), 37, 149-192.
  2. Cherkassky V. & Ma Y. 2005. Multiple Model Regression Estimation, IEEE Transactions on Neural Networks, 16, 785-798.
  3. Cortes, C. and Vapnik, V. 1995. Support-vector networks. Machine Learning, 20, 273-297.
  4. Ferrari-Trecate, G., Muselli, M. 2002. A new learning method for piecewise linear regression In: International Conference on Artificial Neural Networks, Springer.
  5. Fischer, M. 2008. Modeling and forecasting energy demand: Principles and difficulties. In: Proceedings of the NATO Advanced Research Workshop on Weather/Climate Risk Management for the Energy Sector, 207-226.
  6. Guthery, S. B. 1974. Partition regression. Journal of the American Statistical Association, 69, 945-947.
  7. Hathaway, R. J. & Bezdek, J. C. 1993. Switching regression models and fuzzy clustering. IEEE Transactions on Fuzzy Systems 1, 195-204.
  8. Honkela, A. 2001. Nonlinear Switching State-Space Models, Master's Thesis, Helsinki University of Technology, Dep. of Engineering Physics and Mathematics.
  9. Höppner, F. & Klawonn, F. 2003. Improved fuzzy partitions for fuzzy regression models. International Journal of Approximate Reasoning 32, 85-102.
  10. Jahandideh, S. et al. 2009. The use of artificial neural networks and multiple linear regression to predict rate of medical waste generation. Waste Management, 21, 2874-2879.
  11. Kalaba, R., Rasakhoo, N. & Tesfatsion, L. 1989. A FORTRAN program for time-varying linear regression via flexible least squares. Computational Statistics & Data Analysis, 7, 291-309.
  12. Kuchenhof, H. 1996. An exact algorithm for estimating breakpoints in segmented generalized linear models. University of Munich.
  13. Loesch, D. Z., et al. 2006. Transcript levels of the intermediate size or grey zone fragile X mental retardation 1 alleles are raised, and correlate with the number of CGG repeats. Journal of medical genetics, 44.
  14. Nusser, S., Otte, C. & Hauptmann, W. 2008. An EMBased Piecewise Linear Regression Algorithm. Lecture notes in computer science, 466-474.
  15. Niknafs, A., Sun, B., Richter, M. & Ruhe, G. 2011. Predictions in Time Series with Repeated Patterns, Using Piecewise Linear Regression, Technical Report SEDS-TR-094/2011, University of Calgary.
  16. RAPID-I, http://rapid-i.com/content/view/60/200/, accessed at December 2010.
  17. Sikka, G., Kaur, A. & Uddin, M. 2010. Estimating Function points: Using Machine Learning and Regression Models. In: 2nd International Conforence on Education Technology and Computer, 52-56.
Download


Paper Citation


in Harvard Style

Niknafs A., Sun B., M. Richter M. and Ruhe G. (2011). COMPARATIVE ANALYSIS OF THREE TECHNIQUES FOR PREDICTIONS IN TIME SERIES HAVING REPETITIVE PATTERNS . In Proceedings of the 13th International Conference on Enterprise Information Systems - Volume 1: ICEIS, ISBN 978-989-8425-53-9, pages 177-182. DOI: 10.5220/0003463601770182


in Bibtex Style

@conference{iceis11,
author={Arash Niknafs and Bo Sun and Michael M. Richter and Günther Ruhe},
title={COMPARATIVE ANALYSIS OF THREE TECHNIQUES FOR PREDICTIONS IN TIME SERIES HAVING REPETITIVE PATTERNS },
booktitle={Proceedings of the 13th International Conference on Enterprise Information Systems - Volume 1: ICEIS,},
year={2011},
pages={177-182},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003463601770182},
isbn={978-989-8425-53-9},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 13th International Conference on Enterprise Information Systems - Volume 1: ICEIS,
TI - COMPARATIVE ANALYSIS OF THREE TECHNIQUES FOR PREDICTIONS IN TIME SERIES HAVING REPETITIVE PATTERNS
SN - 978-989-8425-53-9
AU - Niknafs A.
AU - Sun B.
AU - M. Richter M.
AU - Ruhe G.
PY - 2011
SP - 177
EP - 182
DO - 10.5220/0003463601770182