Efficient robust nonparametric estimation in a semimartingale regression model

Victor Konev, Serguei Pergamenshchikov

    Research output: Contribution to journalArticle

    5 Citations (Scopus)

    Abstract

    The paper considers the problem of robust estimating a periodic function in a continuous time regression model with the dependent disturbances given by a general square integrable semimartingale with an unknown distribution. An example of such a noise is a non-Gaussian Ornstein-Uhlenbeck process with jumps (see (J. R. Stat. Soc. Ser. B Stat. Methodol. 63 (2001) 167-241), (Ann. Appl. Probab. 18 (2008) 879-908)). An adaptive model selection procedure, based on the weighted least square estimates, is proposed. Under general moment conditions on the noise distribution, sharp non-asymptotic oracle inequalities for the robust risks have been derived and the robust efficiency of the model selection procedure has been shown. It is established that, in the case of the non-Gaussian Ornstein-Uhlenbeck noise, the sharp lower bound for the robust quadratic risk is determined by the limit value of the noise intensity at high frequencies. An example with a martinagale noise exhibits that the risk convergence rate becomes worse if the noise intensity is unbounded.

    Original languageEnglish
    Pages (from-to)1217-1244
    Number of pages28
    JournalAnnales de l'institut Henri Poincare (B) Probability and Statistics
    Volume48
    Issue number4
    DOIs
    Publication statusPublished - Nov 2012

    Fingerprint

    Semimartingale
    Robust Estimation
    Nonparametric Estimation
    Regression Model
    Selection Procedures
    Model Selection
    Oracle Inequalities
    Weighted Estimates
    Least Squares Estimate
    Continuous-time Model
    Moment Conditions
    Ornstein-Uhlenbeck Process
    Weighted Least Squares
    Periodic Functions
    Convergence Rate
    Jump
    Disturbance
    Lower bound
    Unknown
    Nonparametric estimation

    Keywords

    • Asymptotic efficiency
    • Model selection
    • Non-asymptotic estimation
    • Robust risk
    • Sharp oracle inequality

    ASJC Scopus subject areas

    • Statistics and Probability
    • Statistics, Probability and Uncertainty

    Cite this

    Efficient robust nonparametric estimation in a semimartingale regression model. / Konev, Victor; Pergamenshchikov, Serguei.

    In: Annales de l'institut Henri Poincare (B) Probability and Statistics, Vol. 48, No. 4, 11.2012, p. 1217-1244.

    Research output: Contribution to journalArticle

    @article{9828988f14604d63ab3b29c349be5e98,
    title = "Efficient robust nonparametric estimation in a semimartingale regression model",
    abstract = "The paper considers the problem of robust estimating a periodic function in a continuous time regression model with the dependent disturbances given by a general square integrable semimartingale with an unknown distribution. An example of such a noise is a non-Gaussian Ornstein-Uhlenbeck process with jumps (see (J. R. Stat. Soc. Ser. B Stat. Methodol. 63 (2001) 167-241), (Ann. Appl. Probab. 18 (2008) 879-908)). An adaptive model selection procedure, based on the weighted least square estimates, is proposed. Under general moment conditions on the noise distribution, sharp non-asymptotic oracle inequalities for the robust risks have been derived and the robust efficiency of the model selection procedure has been shown. It is established that, in the case of the non-Gaussian Ornstein-Uhlenbeck noise, the sharp lower bound for the robust quadratic risk is determined by the limit value of the noise intensity at high frequencies. An example with a martinagale noise exhibits that the risk convergence rate becomes worse if the noise intensity is unbounded.",
    keywords = "Asymptotic efficiency, Model selection, Non-asymptotic estimation, Robust risk, Sharp oracle inequality",
    author = "Victor Konev and Serguei Pergamenshchikov",
    year = "2012",
    month = "11",
    doi = "10.1214/12-AIHP488",
    language = "English",
    volume = "48",
    pages = "1217--1244",
    journal = "Annales de l'institut Henri Poincare (B) Probability and Statistics",
    issn = "0246-0203",
    publisher = "Institute of Mathematical Statistics",
    number = "4",

    }

    TY - JOUR

    T1 - Efficient robust nonparametric estimation in a semimartingale regression model

    AU - Konev, Victor

    AU - Pergamenshchikov, Serguei

    PY - 2012/11

    Y1 - 2012/11

    N2 - The paper considers the problem of robust estimating a periodic function in a continuous time regression model with the dependent disturbances given by a general square integrable semimartingale with an unknown distribution. An example of such a noise is a non-Gaussian Ornstein-Uhlenbeck process with jumps (see (J. R. Stat. Soc. Ser. B Stat. Methodol. 63 (2001) 167-241), (Ann. Appl. Probab. 18 (2008) 879-908)). An adaptive model selection procedure, based on the weighted least square estimates, is proposed. Under general moment conditions on the noise distribution, sharp non-asymptotic oracle inequalities for the robust risks have been derived and the robust efficiency of the model selection procedure has been shown. It is established that, in the case of the non-Gaussian Ornstein-Uhlenbeck noise, the sharp lower bound for the robust quadratic risk is determined by the limit value of the noise intensity at high frequencies. An example with a martinagale noise exhibits that the risk convergence rate becomes worse if the noise intensity is unbounded.

    AB - The paper considers the problem of robust estimating a periodic function in a continuous time regression model with the dependent disturbances given by a general square integrable semimartingale with an unknown distribution. An example of such a noise is a non-Gaussian Ornstein-Uhlenbeck process with jumps (see (J. R. Stat. Soc. Ser. B Stat. Methodol. 63 (2001) 167-241), (Ann. Appl. Probab. 18 (2008) 879-908)). An adaptive model selection procedure, based on the weighted least square estimates, is proposed. Under general moment conditions on the noise distribution, sharp non-asymptotic oracle inequalities for the robust risks have been derived and the robust efficiency of the model selection procedure has been shown. It is established that, in the case of the non-Gaussian Ornstein-Uhlenbeck noise, the sharp lower bound for the robust quadratic risk is determined by the limit value of the noise intensity at high frequencies. An example with a martinagale noise exhibits that the risk convergence rate becomes worse if the noise intensity is unbounded.

    KW - Asymptotic efficiency

    KW - Model selection

    KW - Non-asymptotic estimation

    KW - Robust risk

    KW - Sharp oracle inequality

    UR - http://www.scopus.com/inward/record.url?scp=84879249535&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84879249535&partnerID=8YFLogxK

    U2 - 10.1214/12-AIHP488

    DO - 10.1214/12-AIHP488

    M3 - Article

    VL - 48

    SP - 1217

    EP - 1244

    JO - Annales de l'institut Henri Poincare (B) Probability and Statistics

    JF - Annales de l'institut Henri Poincare (B) Probability and Statistics

    SN - 0246-0203

    IS - 4

    ER -