Influence diagnostics in a vector autoregressive model

Yonghui Liu, Guocheng Ji, Shuangzhe LIU

    Research output: Contribution to journalArticle

    6 Citations (Scopus)
    6 Downloads (Pure)

    Abstract

    In this paper, we use a likelihood approach and the local influence method introduced by Cook [Assessment of local influence (with discussion). J Roy Statist Soc Ser B. 1986;48:133–149] to study a vector autoregressive (VAR) model. We present the maximum likelihood estimators and the information matrix. We establish the normal curvature and slope diagnostics for the VAR model under several perturbation schemes and use the Monte Carlo method to obtain benchmark values for determining the influence of directional diagnostics and possible influential observations. An empirical study using the VAR model to fit real data of monthly returns of IBM and SP500 index illustrates the effectiveness of our proposed diagnostics.
    Original languageEnglish
    Pages (from-to)2632-2655
    Number of pages24
    JournalJournal of Statistical Computation and Simulation
    Volume85
    Issue number13
    DOIs
    Publication statusPublished - 2015

    Fingerprint

    Influence Diagnostics
    Vector Autoregressive Model
    Local Influence
    Diagnostics
    Normal Curvature
    Influential Observations
    Information Matrix
    Maximum Likelihood Estimator
    Monte Carlo method
    Empirical Study
    Maximum likelihood
    Likelihood
    Slope
    Monte Carlo methods
    Benchmark
    Perturbation
    Vector autoregressive model

    Cite this

    @article{e81f0e4b2fc447e1b45e74abce5fcfb7,
    title = "Influence diagnostics in a vector autoregressive model",
    abstract = "In this paper, we use a likelihood approach and the local influence method introduced by Cook [Assessment of local influence (with discussion). J Roy Statist Soc Ser B. 1986;48:133–149] to study a vector autoregressive (VAR) model. We present the maximum likelihood estimators and the information matrix. We establish the normal curvature and slope diagnostics for the VAR model under several perturbation schemes and use the Monte Carlo method to obtain benchmark values for determining the influence of directional diagnostics and possible influential observations. An empirical study using the VAR model to fit real data of monthly returns of IBM and SP500 index illustrates the effectiveness of our proposed diagnostics.",
    keywords = "maximum-likelihood-estimator, perturbation-scheme, curvature",
    author = "Yonghui Liu and Guocheng Ji and Shuangzhe LIU",
    year = "2015",
    doi = "10.1080/00949655.2014.967243",
    language = "English",
    volume = "85",
    pages = "2632--2655",
    journal = "Journal of Statistical Computation and Simulation",
    issn = "0094-9655",
    publisher = "Taylor and Francis Ltd.",
    number = "13",

    }

    Influence diagnostics in a vector autoregressive model. / Liu, Yonghui; Ji, Guocheng; LIU, Shuangzhe.

    In: Journal of Statistical Computation and Simulation, Vol. 85, No. 13, 2015, p. 2632-2655.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Influence diagnostics in a vector autoregressive model

    AU - Liu, Yonghui

    AU - Ji, Guocheng

    AU - LIU, Shuangzhe

    PY - 2015

    Y1 - 2015

    N2 - In this paper, we use a likelihood approach and the local influence method introduced by Cook [Assessment of local influence (with discussion). J Roy Statist Soc Ser B. 1986;48:133–149] to study a vector autoregressive (VAR) model. We present the maximum likelihood estimators and the information matrix. We establish the normal curvature and slope diagnostics for the VAR model under several perturbation schemes and use the Monte Carlo method to obtain benchmark values for determining the influence of directional diagnostics and possible influential observations. An empirical study using the VAR model to fit real data of monthly returns of IBM and SP500 index illustrates the effectiveness of our proposed diagnostics.

    AB - In this paper, we use a likelihood approach and the local influence method introduced by Cook [Assessment of local influence (with discussion). J Roy Statist Soc Ser B. 1986;48:133–149] to study a vector autoregressive (VAR) model. We present the maximum likelihood estimators and the information matrix. We establish the normal curvature and slope diagnostics for the VAR model under several perturbation schemes and use the Monte Carlo method to obtain benchmark values for determining the influence of directional diagnostics and possible influential observations. An empirical study using the VAR model to fit real data of monthly returns of IBM and SP500 index illustrates the effectiveness of our proposed diagnostics.

    KW - maximum-likelihood-estimator

    KW - perturbation-scheme

    KW - curvature

    U2 - 10.1080/00949655.2014.967243

    DO - 10.1080/00949655.2014.967243

    M3 - Article

    VL - 85

    SP - 2632

    EP - 2655

    JO - Journal of Statistical Computation and Simulation

    JF - Journal of Statistical Computation and Simulation

    SN - 0094-9655

    IS - 13

    ER -