Convergence analysis of ladder algorithms for AR and ARMA models (Q1075991)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Convergence analysis of ladder algorithms for AR and ARMA models
scientific article

    Statements

    Convergence analysis of ladder algorithms for AR and ARMA models (English)
    0 references
    0 references
    0 references
    0 references
    0 references
    1986
    0 references
    The paper establishes the convergence properties of the exact recursive least-squares ladder/lattice algorithm for AR models, and of its extension (in the manner of the extended least-squares (ELS) methods) to ARMA models. The ordinary differential equation (ODE) approach is used to analyse the two algorithms. As expected, the AR algorithm is shown to be convergent under weak conditions. The ARMA algorithm is shown to converge under a positive realness condition on the MA part of the ARMA system (this again being somewhat expected in view of the analogy with ELS).
    0 references
    autoregressive models
    0 references
    autoregressive moving-average models
    0 references
    convergence properties
    0 references
    exact recursive least-squares ladder/lattice algorithm
    0 references
    positive realness
    0 references
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references