A test for the homoscedasticity of the residuals in fuzzy rule-based forecasters
- 81 Downloads
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.
KeywordsFuzzy rule-based systems Heteroscedasticity Residuals Diagnostic checking
Unable to display preview. Download preview PDF.
- 1.Aznarte JL (2008) Modelling time series through fuzzy rule-based models: a statistical approach. PhD thesis, Universidad de Granada Google Scholar
- 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
- 9.Barreto H, Howland F (2005) Introductory econometrics: using Monte Carlo simulation with Microsoft excel. Cambridge University Press, Cambridge Google Scholar
- 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. http://www.sciencedirect.com/science/article/B6V0T-4NFH0BH-2/2/162d0452aebb66d4e95da94e1e1e842b. Mathematical aspects of modelling and control MathSciNetzbMATHCrossRefGoogle Scholar
- 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.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
- 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. http://ideas.repec.org/a/bes/jnlasa/v99y2004p1092-1107.html MathSciNetzbMATHCrossRefGoogle Scholar
- 30.Tong H (1978) On a threshold model. In: Pattern recognition and signal processing Google Scholar
- 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.Wang LX (1992) Fuzzy systems are universal approximators. IEEE Trans Syst Man Cybern SMC7(10):1163–1170 Google Scholar