Model structure selection for multivariable systems by cross-validation methods

  title={Model structure selection for multivariable systems by cross-validation methods},
  author={P.H.M. Janssen and P. Stoica and T. S{\"o}derstr{\"o}m},
  • P.H.M. Janssen, P. Stoica, T. Söderström
  • Published 1987
  • Mathematics, Computer Science
  • Using cross-validation ideas, two procedures are proposed for making a choice between different model structures used for (approximate) modelling of multivariable systems. The procedures are derived under fairly general conditions: the ‘true’ system does not need to be contained in the model set; model structures do not need to be nested and different criteria may be used for model estimation and validation. The proposed structure selection rules are shown to be invariant to parameter scaling… CONTINUE READING
    26 Citations
    Parametric uncertainty in system identification
    • 14
    • PDF
    Ten iterative steps in development and evaluation of environmental models
    • 925
    • PDF
    A combined neural network and decision trees model for prognosis of breast cancer relapse
    • 197
    • PDF
    Bankruptcy prediction of financially stressed firms: an examination of the predictive accuracy of artificial neural networks
    • 108
    Evolving Modular Neural Networks
    • W. Liu, Xin YaoComputational
    • 1997
    • 4
    Evolving modular neural networks which generalise well
    • Yong Liu, X. Yao
    • Computer Science
    • Proceedings of 1997 IEEE International Conference on Evolutionary Computation (ICEC '97)
    • 1997
    • 32


    Model-structure selection by cross-validation
    • 123
    • PDF
    Convergence analysis of parametric identification methods
    • 435
    • Highly Influential
    On-line structure selection for multivariable state-space models
    • 90
    Asymptotic normality of prediction error estimators for approximate system models
    • 158
    A new look at the statistical model identification
    • 36,856
    • PDF
    On the choice of parameterization for identification
    • 8
    Optimal experiment designs with respect to the intended model application
    • 286
    • 77