Advertisement

Linear regression in a fuzzy context. The least square method

  • Antonia Salas
  • Norberto Corral
  • Carlo Bertoluzza
Part of the Studies in Fuzziness and Soft Computing book series (STUDFUZZ, volume 87)

Abstract

This paper deals with the statistical linear regression. Its aim is to extend the classical least square method to the case where the observations are not crisp, but fuzzy numbers. In order to attain our purpose, we introduce a suitable and very general squared distance between fuzzy numbers which substitute the classical (ab)2 on the real line. Then we use a well known theorem of functional theory to prove that, under suitable and reasonable conditions, there exists a unique set of fuzzy coefficients which minimize the sum of the squared distances between the previsions and the observations. Nevertheless the classical variational methods cannot be used to find the regression coefficient due to the particular structure of the domain of the functional to be minimized (its interior is empty). So we complete the paper by describing a numerical solution based on active constraint method.

Keywords

Fuzzy Number Fuzzy Subset Active Constraint Unique Minimum Constraint Matrix 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Baiocchi, C. and Capelo, A. (1984). Variational and Quasivariational inequalities, John Wiley and Sons, New York.MATHGoogle Scholar
  2. 2.
    Bertoluzza, C., Corral, N. and Salas, A. (1991). Fuzzy Linear Regression: Existence of Solution for a generalized Least Square Method, Proc. IFSA’91, Bruxelles.Google Scholar
  3. 3.
    Bertoluzza, C., Corral, N. and Salas, A. (1995). On a new class of distances between fuzzy sets, Mathware fi Soft Computing 2, 71–84.MathSciNetMATHGoogle Scholar
  4. 4.
    Bollani, C. (2001). Determinazione approssimata dei coefficienti di regressione in ambiente sfumato (temptative title). Tesi di Laurea, Università di Pavia (to be discussed).Google Scholar
  5. 5.
    Ciampa, S. (1991–92). Problemi connessi con la definizione di retta sfumata,Tesi di Laurea, Università di Pavia.Google Scholar
  6. 6.
    Ciliberti, M.G. (1990–91). Una possibile definizione di distanza per punti sfumati,Tesi di Laurea, Università di Pavia.Google Scholar
  7. 7.
    Diamond, P. (1991). Least Square Method in Fuzzy Data Analysis, Proc. IFSA’91, Bruxelles.Google Scholar
  8. 8.
    Gill, P.E., Murray, W. and Wright, M.H. (1981). Practical Optimization, Academic Press, London.MATHGoogle Scholar
  9. 9.
    Kaufmann, A. and Gupta, M.M. (1985). Introduction to Fuzzy Arithmetics; Theory and Applications, Van Nostrand Reinhold, New York.Google Scholar
  10. 10.
    Mainardi, S. (1991–92). Regressione lineare sfumata: un’analisi dei parametri che influenzano i coefficienti di interpolazione,Tesi di Laurea, Università di Pavia.Google Scholar
  11. 11.
    Nguyen, H.T. (1978). A Note on the Extension Principle for Fuzzy Sets, J. Math. Anal. Appl. 64, 369–380.MathSciNetMATHCrossRefGoogle Scholar
  12. 12.
    Salas, A. (1991). Regresión lineal con observaciones difusas, PhD Thesis, Universidad de Oviedo.Google Scholar
  13. 13.
    Zadeh, L.A. (1965). Fuzzy sets, Information and Control 8, 338–353.MathSciNetMATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Antonia Salas
    • 1
  • Norberto Corral
    • 1
  • Carlo Bertoluzza
    • 2
  1. 1.Departamento de Estadística e I.O. y D.M.Universidad de Oviedo, Facultad de CienciasOviedoSpain
  2. 2.Dipartimento di Informatica e SistemisticaUniversity of PaviaPaviaItaly

Personalised recommendations