Truncated sequential estimation of the parameter of a first order autoregressive process with dependent noises

D. Fourdrinier, V. Konev, S. Pergamenshchikov

    Research output: Contribution to journalArticle

    6 Citations (Scopus)

    Abstract

    For a first-order non-explosive autoregressive process with dependent noise, we propose a truncated sequential procedure with a fixed mean-square accuracy. The asymptotic distribution of the estimator depends on the type of the noise distribution: it is normal when the noise has a Kotz's distribution, while it is a mixture of normal distributions if the noise distribution is a variance mixture of normal distrbutions as well. In both cases, the convergence to the limiting distribution is uniform in the unknown parameter.

    Original languageEnglish
    Pages (from-to)43-58
    Number of pages16
    JournalMathematical Methods of Statistics
    Volume18
    Issue number1
    DOIs
    Publication statusPublished - 1 Mar 2009

    Fingerprint

    Sequential Estimation
    Autoregressive Process
    First-order
    Dependent
    Mixture of Normal Distributions
    Sequential Procedure
    Limiting Distribution
    Unknown Parameters
    Mean Square
    Asymptotic distribution
    Estimator
    Autoregressive process

    Keywords

    • autoregression model
    • truncated sequential estimators
    • uniform normality

    ASJC Scopus subject areas

    • Statistics, Probability and Uncertainty
    • Statistics and Probability

    Cite this

    Truncated sequential estimation of the parameter of a first order autoregressive process with dependent noises. / Fourdrinier, D.; Konev, V.; Pergamenshchikov, S.

    In: Mathematical Methods of Statistics, Vol. 18, No. 1, 01.03.2009, p. 43-58.

    Research output: Contribution to journalArticle

    @article{6248b74670fe4a41ad59f5157f21ff59,
    title = "Truncated sequential estimation of the parameter of a first order autoregressive process with dependent noises",
    abstract = "For a first-order non-explosive autoregressive process with dependent noise, we propose a truncated sequential procedure with a fixed mean-square accuracy. The asymptotic distribution of the estimator depends on the type of the noise distribution: it is normal when the noise has a Kotz's distribution, while it is a mixture of normal distributions if the noise distribution is a variance mixture of normal distrbutions as well. In both cases, the convergence to the limiting distribution is uniform in the unknown parameter.",
    keywords = "autoregression model, truncated sequential estimators, uniform normality",
    author = "D. Fourdrinier and V. Konev and S. Pergamenshchikov",
    year = "2009",
    month = "3",
    day = "1",
    doi = "10.3103/S1066530709010037",
    language = "English",
    volume = "18",
    pages = "43--58",
    journal = "Mathematical Methods of Statistics",
    issn = "1066-5307",
    publisher = "Allerton Press Inc.",
    number = "1",

    }

    TY - JOUR

    T1 - Truncated sequential estimation of the parameter of a first order autoregressive process with dependent noises

    AU - Fourdrinier, D.

    AU - Konev, V.

    AU - Pergamenshchikov, S.

    PY - 2009/3/1

    Y1 - 2009/3/1

    N2 - For a first-order non-explosive autoregressive process with dependent noise, we propose a truncated sequential procedure with a fixed mean-square accuracy. The asymptotic distribution of the estimator depends on the type of the noise distribution: it is normal when the noise has a Kotz's distribution, while it is a mixture of normal distributions if the noise distribution is a variance mixture of normal distrbutions as well. In both cases, the convergence to the limiting distribution is uniform in the unknown parameter.

    AB - For a first-order non-explosive autoregressive process with dependent noise, we propose a truncated sequential procedure with a fixed mean-square accuracy. The asymptotic distribution of the estimator depends on the type of the noise distribution: it is normal when the noise has a Kotz's distribution, while it is a mixture of normal distributions if the noise distribution is a variance mixture of normal distrbutions as well. In both cases, the convergence to the limiting distribution is uniform in the unknown parameter.

    KW - autoregression model

    KW - truncated sequential estimators

    KW - uniform normality

    UR - http://www.scopus.com/inward/record.url?scp=84859509830&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84859509830&partnerID=8YFLogxK

    U2 - 10.3103/S1066530709010037

    DO - 10.3103/S1066530709010037

    M3 - Article

    VL - 18

    SP - 43

    EP - 58

    JO - Mathematical Methods of Statistics

    JF - Mathematical Methods of Statistics

    SN - 1066-5307

    IS - 1

    ER -