Journal of Global Optimization

, Volume 60, Issue 1, pp 103–120 | Cite as

Efficient adaptive regression spline algorithms based on mapping approach with a case study on finance

  • Elcin Kartal Koc
  • Cem Iyigun
  • İnci Batmaz
  • Gerhard-Wilhelm Weber


Multivariate adaptive regression splines (MARS) has become a popular data mining (DM) tool due to its flexible model building strategy for high dimensional data. Compared to well-known others, it performs better in many areas such as finance, informatics, technology and science. Many studies have been conducted on improving its performance. For this purpose, an alternative backward stepwise algorithm is proposed through Conic-MARS (CMARS) method which uses a penalized residual sum of squares for MARS as a Tikhonov regularization problem. Additionally, by modifying the forward step of MARS via mapping approach, a time efficient procedure has been introduced by S-FMARS. Inspiring from the advantages of MARS, CMARS and S-FMARS, two hybrid methods are proposed in this study, aiming to produce time efficient DM tools without degrading their performances especially for large datasets. The resulting methods, called SMARS and SCMARS, are tested in terms of several performance criteria such as accuracy, complexity, stability and robustness via simulated and real life datasets. As a DM application, the hybrid methods are also applied to an important field of finance for predicting interest rates offered by a Turkish bank to its customers. The results show that the proposed hybrid methods, being the most time efficient with competing performances, can be considered as powerful choices particularly for large datasets.


Regression splines Optimization Multivariate adaptive regression splines (MARS) CMARS Self organizing maps Data mining Interest rate estimation 



Authors would like to acknowledge one of the well-known Turkish banks for the financial data provision. They would like to thank CeydaYazıcı for the preparation of the financial data, and Dr. Seza Danışoğlu of METU, for her contributions to the financial application section.


  1. 1.
    Austin, P.C.: A comparison of regression trees, logistic regression, generalized additive models and multivariate adaptive regression splines for predicting. AMI Mortal. Stat. Med. 26(15), 2937–2957 (2007)CrossRefGoogle Scholar
  2. 2.
    Batmaz, İ., Yerlikaya-Özkurt, F., Kartal-Koc, E., Köksal, G., Weber, G.-W.: Evaluating the CMARS performance for modeling nonlinearities. In: American Institute of Physics Conference Proceedings, Power Control and Optimization: Proceedings of the 3rd Global Conference on Power Control and Optimization, Gold Coast, Australia, 2–4, vol. 1239, pp. 351–357 (2010)Google Scholar
  3. 3.
    Batmaz, İ., Kartal-Koç, E., Yazıcı, C.: Comparison of parametric and nonparametric statistical models for predicting interest rates. In: EURO/INFORMS: The 26th European Conference on Operational Research, 1–4 July, Rome, Abstract Book, p. 123 (2013)Google Scholar
  4. 4.
    Chou, S.M., Lee, T.S., Shao, Y.E., Chen, I.F.: Mining the breast cancer pattern using artificial neural networks and multivariate adaptive regression splines. Exp. Syst. Appl. 27, 133–142 (2004)CrossRefGoogle Scholar
  5. 5.
    Deconinck, E., Coomons, D., Heyden, Y.V.: Explorations of linear modeling techniques and their combinations with multivariate adaptive regression splines to predict gastro-intestinal absorption of drugs. J. Pharm. Biomed. Anal. 43(1), 119–130 (2007)CrossRefGoogle Scholar
  6. 6.
    Denison, D.G.T., Mallick, B.K., Smith, A.F.M.: Automatic bayesian curve fitting. J. R. Stat. Soc. 60, 333–350 (1998)CrossRefGoogle Scholar
  7. 7.
    DiMatteo, I., Genovese, C.R., Kass, R.E.: Bayesian curve fitting with free-knot splines. Biometrika 88, 1055–1071 (2001)CrossRefGoogle Scholar
  8. 8.
    Friedman, J.H., Silverman, B.W.: Flexible parsimonious smoothing and additive modelling. Technometrics 31, 3–21 (1989)CrossRefGoogle Scholar
  9. 9.
    Friedman, J.H.: Multivariate adaptive regression splines. Ann. Stat. 19, 1–67 (1991)CrossRefGoogle Scholar
  10. 10.
    Friedman, J.H.: Fast MARS. Technical Report No: 110, Stanford University Department of Statistics (1993)Google Scholar
  11. 11.
    Haas, H., Kubin, G.: A multi-band nonlinear oscillator model for speech. In: Conference Record of the Thirty-Second Asilomar Conference on Signals, Systems and Computers, vol. 1, pp. 338–342 (1998)Google Scholar
  12. 12.
    Jekabsons, G.: ARESLab: Adaptive regression splines toolbox for matlab/octave. (2011)
  13. 13.
    Kartal-Koc, E.: an algorithm for the forward step of adaptive regression splines via mapping approach. Ph.D. thesis, Middle East Technical University, Ankara, Turkey (2012)Google Scholar
  14. 14.
    Kartal-Koc, E., Iyigün, C.: Restructuring forward step of mars algorithm using a new knot selection procedure based on a mapping approach. J. Glob. Optim. doi: 10.1007/s10898-013-0107-5
  15. 15.
    Kohonen, T.: Self-organizing and Associative Memory. Springer, New York (1988)CrossRefGoogle Scholar
  16. 16.
    Kubin, G.: Nonlinear prediction of mobile radio channels: measurments and mars model designs. In: IEEE Proceedings International Conference on Acoustics, Speech, and Signal Processing, 5, 15–19, pp. 2667–2670 (1999)Google Scholar
  17. 17.
    Lee, T.S., Chiu, C.C., Chou, Y.C., Lu, C.J.: Mining the customer credit using classification and regression tree and multivariate adaptive regression splines. Comput. Stat. Data Anal. 50, 1113–1130 (2006)CrossRefGoogle Scholar
  18. 18.
    Lou, Z., Wahba, G.: Hybrid adaptive splines. J. Am. Stat. Assoc. 92, 107–116 (1997)CrossRefGoogle Scholar
  19. 19.
    MOSEK, A very powerful commercial software for CQP.
  20. 20.
    Nesterov, Y.E., Nemirovski, A.S.: Interior Point Polynomial Algorithms in Convex Programming. SIAM Publications, Philadelphia (1994)CrossRefGoogle Scholar
  21. 21.
    Özmen, A.: Robust conic quadratic programming applied to quality improvement-a robustification of CMARS. MSc Thesis, Middle East Technical University, Ankara, Turkey (2010)Google Scholar
  22. 22.
    Özmen, A., Weber, G.-W., Batmaz, İ., Kropat, E.: RCMARS: Robustification of CMARS with different scenarios under polyhedral uncertainty set. In: Communications in Nonlinear Science and Numerical Simulation (CNSNS): Nonlinear, Fractional and Complex, vol. 16, pp. 4780–4787 (2011)Google Scholar
  23. 23.
    Smith, P.L.: Curve fitting and modeling with splines using statistical variable selection techniques. Report NASA 166034, NASA, Langley Research Center, Hampton (1982)Google Scholar
  24. 24.
    Stone, C., Hansen, M., Kooperberg, C., Troung, Y.: Polynomial splines and their tensor products in extended linear modeling. Ann. Stat. 25, 1371–1470 (1997)CrossRefGoogle Scholar
  25. 25.
    Taylan, P., Weber, G.W., Beck, A.: New approaches to regression by generalized additive models and continuous optimization for modern applications in finance, science and technology. J. Optim. 56, 675–698 (2007)CrossRefGoogle Scholar
  26. 26.
    Vesanto, J., Himberg, J., Alhoniehmi, E.: Parhankangas, SOM toolbox for matlab 5, report A57. (2000)
  27. 27.
    Weber, G.W., Batmaz, İ., Köksal, G., Taylan, P., Yerlikaya-Özkurt, F.: CMARS: A new contribution to nonparametric regression with multivariate adaptive regression splines supported by continuous optimization. Inverse Probl. Sci. Eng. 20(3), 371–400 (2012)CrossRefGoogle Scholar
  28. 28.
    Yazıcı, C.: A computational approach to nonparametric regression: bootstrapping the CMARS Method. MSc Thesis, Middle East Technical University, Ankara, Turkey (2011)Google Scholar
  29. 29.
    Yazıcı, C., Yerlikaya-Özkurt, F., Batmaz, İ.: A computational approach to nonparametric regression: bootstrapping the CMARS method. In: CFE-ERCIM 2011: 4th International Conference of the ERCIM (European Research Consortium for Informatics and Mathematics) Working Group on Computing and Statistics, London, UK (2011)Google Scholar
  30. 30.
    Yerlikaya, F.: A new contribution to nonlinear robust regression and classification with MARS and its application to data mining for quality control in manufacturing. MSc Thesis, Graduate school of applied mathematics, Middle East Technical University, Ankara, Turkey (2008)Google Scholar
  31. 31.
    Yerlikaya-Özkurt, F., Batmaz, İ., Weber, G.-W.: A review of conic multivariate adaptive regression splines (CMARS): a powerful tool for predictive data mining. In: Zilberman, D., Pinto, A. (eds.) Springer Volume Modeling, Optimization, Dynamics and Bioeconomy. Series Springer Proceedings in Mathematics (in print) (2012)Google Scholar
  32. 32.
    Zhou, S., Shen, X.: Spatially adaptive regression splines and accurate knot selection schemes. J. Am. Stat. Assoc. 96, 247–259 (2012)CrossRefGoogle Scholar
  33. 33.
    Zhou, Y., Leung, H.: Predicting object-oriented software maintainability using multivariate adaptive regression splines. J. Syst. Softw. 80(8), 1349–1361 (2007)CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2014

Authors and Affiliations

  • Elcin Kartal Koc
    • 1
  • Cem Iyigun
    • 2
  • İnci Batmaz
    • 3
  • Gerhard-Wilhelm Weber
    • 4
  1. 1.Department of Statistics, Operations and Management ScienceThe University of TennesseeKnoxvilleUSA
  2. 2.Department of Industrial EngineeringMiddle East Technical UniversityAnkaraTurkey
  3. 3.Department of StatisticsMiddle East Technical UniversityAnkaraTurkey
  4. 4.Institute of Applied MathematicsMiddle East Technical UniversityAnkaraTurkey

Personalised recommendations