Gauss-Newton and M-estimation for ARMA processes with infinite variance (Q1272156)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Gauss-Newton and M-estimation for ARMA processes with infinite variance
scientific article

    Statements

    Gauss-Newton and M-estimation for ARMA processes with infinite variance (English)
    0 references
    0 references
    23 November 1998
    0 references
    The author considers two estimation procedures, Gauss-Newton and M-estimation, for the parameters of an ARMA(\(p,q\)) process when the innovations belong to the domain of attraction of a non-normal stable distribution. It is shown that the Gauss-Newton and iterative least squares estimators have the same limit distribution as the maximum likelihood and Whittle estimators. The weak convergence for a class of M-estimators is also established. A small simulation study documents the performance of the studied estimators for finite sample sizes.
    0 references
    Gauss-Newton estimate
    0 references
    heavy tails
    0 references
    stable distributions
    0 references
    M-estimation
    0 references
    ARMA processes
    0 references

    Identifiers