Applied Intelligence

, Volume 34, Issue 3, pp 386–393 | Cite as

A test for the homoscedasticity of the residuals in fuzzy rule-based forecasters

  • José Luis AznarteEmail author
  • Daniel Molina
  • Ana M. Sánchez
  • José M. Benítez


Heteroscedasticy is the property of having a changing variance throughout the time. Homoscedasticity is the converse, that is, having a constant variance. This is a key property for time series models which may have serious consequences when making inferences out of the errors of a given forecaster. Thus it has to be conveniently assessed in order to establish the quality of the model and its forecasts. This is important for every model including fuzzy rule-based systems, which have been applied to time series analysis for many years. Lagrange multiplier testing framework is used to evaluate wether the residuals of an FRBS are homoscedastic. The test robustness is thoroughly evaluated through an extensive experimentation. This is another important step towards a statistically sound modeling strategy for fuzzy rule-based systems.


Fuzzy rule-based systems Heteroscedasticity Residuals Diagnostic checking 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Aznarte JL (2008) Modelling time series through fuzzy rule-based models: a statistical approach. PhD thesis, Universidad de Granada Google Scholar
  2. 2.
    Aznarte JL, Benítez Sánchez JM (2006) On the identifiability of TSK additive fuzzy rule-based models. In: Soft methods for integrated uncertainty modelling. Advances in soft computing, vol 6. Springer, Berlin, pp 79–86 Google Scholar
  3. 3.
    Aznarte JL, Benítez JM (2010) Equivalences between neural-autoregressive time series models and fuzzy systems. IEEE Trans Neural Netw 21(9):1434–1444. doi: 10.1109/TNN.2010.2060209 CrossRefGoogle Scholar
  4. 4.
    Aznarte JL, Benítez JM, Castro JL (2007) Smooth transition autoregressive models and fuzzy rule-based systems: Functional equivalence and consequences. Fuzzy Sets Syst 158(24):2734–2745. doi: 10.1016/j.fss.2007.03.021 zbMATHCrossRefGoogle Scholar
  5. 5.
    Aznarte JL, Benítez JM, Nieto-Lugilde D, de Linares Fernández C, Díaz de la Guardia C, Alba Sánchez F (2007) Forecasting airborne pollen concentration time series with neural and neuro-fuzzy models. Expert Syst Appl 32(4):1218–1225. doi: 10.1016/j.eswa.2006.02.011 CrossRefGoogle Scholar
  6. 6.
    Aznarte JL, Medeiros MC, Benítez JM (2010) Linearity testing for fuzzy rule-based models. Fuzzy Sets Syst 161(13):1836–1851 zbMATHCrossRefGoogle Scholar
  7. 7.
    Aznarte JL, Medeiros MC, Benítez JM (2010) Testing for remaining autocorrelation of the residuals in the framework of fuzzy rule-based time series modelling. Int J Uncertain Fuzziness Knowl-Based Syst 18(4):371–387 zbMATHCrossRefGoogle Scholar
  8. 8.
    Bárdossy A, Duckstein L (1995) Fuzzy rule-based modeling with applications to geophysical, biological, and engineering systems. Systems engineering series. CRC Press, Boca Raton. zbMATHGoogle Scholar
  9. 9.
    Barreto H, Howland F (2005) Introductory econometrics: using Monte Carlo simulation with Microsoft excel. Cambridge University Press, Cambridge Google Scholar
  10. 10.
    Benrejeb M, Sakly A, Othman KB, Borne P (2008) Choice of conjunctive operator of tsk fuzzy systems and stability domain study. Math Comput Simul 76(5–6):410–421. doi: 10.1016/j.matcom.2007.04.003. Mathematical aspects of modelling and control MathSciNetzbMATHCrossRefGoogle Scholar
  11. 11.
    Byun H, Lee K (2005) A decision support system for the selection of a rapid prototyping process using teh modified topsis method. Int J Adv Manuf Technol 26(11–12):1338–1347 CrossRefGoogle Scholar
  12. 12.
    Cárdenas E, Castillo J, Cordón O, Herrera F, Peregrín A (1995) Applicability of t-norms in fuzzy control. Busefal 61:28–36 Google Scholar
  13. 13.
    Cordón O, Herrera F, Peregrin A (1995) T-norms versus implication functions as implication operators in fuzzy control. In: Proceedings of the IFSA 1995 conference, Sao Paulo, Brazil, pp 501–504 Google Scholar
  14. 14.
    Driankov D, Hellendoorn H, Reinfrank M (1996) An introduction to fuzzy control, 2nd edn. Springer, Berlin zbMATHGoogle Scholar
  15. 15.
    Jang JS (1993) ANFIS: Adaptive-network-based fuzzy inference system. IEEE Trans Syst Man Cibern 23(3):665–685 MathSciNetCrossRefGoogle Scholar
  16. 16.
    Jang JS, Sun CT (1993) Predicting chaotic time series with fuzzy if-then rules. In: Second IEEE international conference on fuzzy systems, 1993, vol 2, pp 1079–1084. doi: 10.1109/FUZZY.1993.327364 CrossRefGoogle Scholar
  17. 17.
    John R, Innocent P (2005) Modeling uncertainty in clinical diagnosis using fuzzy logic. IEEE Trans Syst Man Cybern, Part B, Cybern 35(6):1340–1350. doi: 10.1109/TSMCB.2005.855588 CrossRefGoogle Scholar
  18. 18.
    Kim J, Kasabov N (1999) HyFIS: adaptive neuro-fuzzy inference systems and their application to nonlinear dynamical systems. Neural Netw 12:1301–1319 CrossRefGoogle Scholar
  19. 19.
    Lee I, Kosko B, Anderson WF (2005) Modeling gunshot bruises in soft body armor with an adaptive fuzzy system. IEEE Trans Syst Man Cybern, Part B, Cybern 35(6):1374–1390 CrossRefGoogle Scholar
  20. 20.
    Maguire L, Roche B, McGuinnity T, McDaid L (1998) Predicting a chaotic time series using a fuzzy neural network. Inf Sci 112(1–4):125–136 zbMATHCrossRefGoogle Scholar
  21. 21.
    Medeiros M, Veiga A (2003) Diagnostic checking in a flexible nonlinear time series model. J Time Ser Anal 24:461–482 MathSciNetzbMATHCrossRefGoogle Scholar
  22. 22.
    Medeiros M, Veiga A (2005) A flexible coefficient smooth transition time series model. IEEE Trans Neural Netw 16(1):97–113 CrossRefGoogle Scholar
  23. 23.
    Nauck D, Kruse R (1999) Neuro-fuzzy systems for function approximation. Fuzzy Sets Syst 101(2):261–271. doi: 10.1016/S0165-0114(98)00169-9 MathSciNetzbMATHCrossRefGoogle Scholar
  24. 24.
    Nie J (1997) Nonlinear time-series forecasting: a fuzzy-neural approach. Neurocomputing 16(1):63–76 CrossRefGoogle Scholar
  25. 25.
    Palit AK, Popovic D (2005) Computational intelligence in time series forecasting: theory and engineering applications (Advances in industrial control). Springer, New York zbMATHGoogle Scholar
  26. 26.
    Suarez-Farinas M, Pedreira CE, Medeiros MC (2004) Local global neural networks: a new approach for nonlinear time series modeling. J Am Stat Assoc 99:1092–1107. MathSciNetzbMATHCrossRefGoogle Scholar
  27. 27.
    Sugeno M, Kang G (1988) Structure identification of fuzzy model. Fuzzy Sets Syst 28:15–33 MathSciNetzbMATHCrossRefGoogle Scholar
  28. 28.
    Takagi T, Sugeno M (1985) Fuzzy identification of systems and its applications to modeling and control. IEEE Trans Syst Man Cybern 15:116–132 zbMATHGoogle Scholar
  29. 29.
    Teräsvirta T (1994) Specification, estimation and evaluation of smooth transition autoregresive models. J Am Stat Assoc 89:208–218 CrossRefGoogle Scholar
  30. 30.
    Tong H (1978) On a threshold model. In: Pattern recognition and signal processing Google Scholar
  31. 31.
    Vieira CF, Palma LB, da Silva RN (2005) Robust fault diagnosis approach using analytical and knowledge based techniques applied to a water tank system. Int J Eng Intell Syst Electr Eng Commun 13(4):237–244 Google Scholar
  32. 32.
    Wang LX (1992) Fuzzy systems are universal approximators. IEEE Trans Syst Man Cybern SMC7(10):1163–1170 Google Scholar
  33. 33.
    Zadeh L (1965) Fuzzy sets. Inf Control 3(8):338–353 MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2011

Authors and Affiliations

  • José Luis Aznarte
    • 1
    Email author
  • Daniel Molina
    • 2
  • Ana M. Sánchez
    • 3
  • José M. Benítez
    • 4
  1. 1.Dept. of Artificial IntelligenceUNEDMadridSpain
  2. 2.Dept. Computer Languages and SystemsUniversity of CádizCádizSpain
  3. 3.Dept. Software EngineeringUniversity of GranadaGranadaSpain
  4. 4.Dept. of Computational Sciences and A. I., CITIC-UGRUniversidad de GranadaGranadaSpain

Personalised recommendations