Estimation of structure by minimum description length
This paper proves a theorem which shows that a previously derived minimum description length (MDL) estimation criterion can be used for distinguishing between structures in linear models for vector processes. This is done by considering an extreme case where the data are generated by a system falling outside of the linear space spanned by the coordinate system of the fitted models. The central result of the paper is the proof that in this case the inner product in the third term of the optimum description length: \[ -(1/N)\log P(y/\theta)+(k/2N)\log N+(k/2N)\log(\theta,I_{\theta}(y)\theta) \] grows to infinity with N, while it remains bounded if a model with the same structure as the data generating system is selected. The paper also presents a simple algorithm for estimating the best model, including its structure and the number of its parameters.
- A new look at the statistical model identification
- Basis of invariants and canonical forms for linear dynamic systems
- scientific article; zbMATH DE number 3603100 (Why is no real title available?)
- scientific article; zbMATH DE number 3635352 (Why is no real title available?)
- scientific article; zbMATH DE number 3262975 (Why is no real title available?)
- scientific article; zbMATH DE number 3365976 (Why is no real title available?)
- Invariant Description of Linear, Time-Invariant Controllable Systems
- Markovian Representation of Stochastic Processes by Canonical Variables
- Modeling by shortest data description
- On line structure selection for multivariable state-space models
- Parametrizations of linear dynamical systems: Canonical forms and identifiability
- Recursive estimation of mixed autoregressive-moving average order
- Some properties of the parameterization of ARMA systems with unknown order
- The determination of optimum structures for the state space representation of multivariate stochastic processes
- The estimation of the order of an ARMA process
- A note on some model selection criteria
- Parameter estimation for continuous-time models - a survey
- On identification of stable systems and optimal approximation
- Twenty-one ML estimators for model selection
- Order estimation of 2-dimensional complex superimposed exponential signal model using exponentially embedded family (EEF) rule: large sample consistency properties
- A unified balanced approach to multichannel blind deconvolution
- Minimum Description Length Principle for Fat-Tailed Distributions
- On consistency of minimum description length model selection for piecewise autoregressions
- scientific article; zbMATH DE number 193126 (Why is no real title available?)
- Recognition of unnatural patterns in manufacturing processes using the minimum description length criterion
- Construction of nonlinear discrimination function base on the MDL criterion.
- A recursive algorithm for estimating multiple models continuous transfer function with non-uniform sampling
- scientific article; zbMATH DE number 6751371 (Why is no real title available?)
- Structural, Syntactic, and Statistical Pattern Recognition
- Information criteria for model selection
- An information criterion for normal regression estimation
- Uniquely identifiable state-space and ARMA parametrizations for multivariable linear systems
This page was built for publication: Estimation of structure by minimum description length
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q794965)
