Influence diagnostic analysis in the possibly heteroskedastic linear model with exact restrictions

Shuangzhe LIU, Victor Leiva, Tiefeng Ma

    Research output: Contribution to journalArticle

    8 Citations (Scopus)

    Abstract

    The local influence method has proven to be a useful and powerful tool for detecting influential observations on the estimation of model parameters. This method has been widely applied in different studies related to econometric and statistical modelling. We propose a methodology based on the Lagrange multiplier method with a linear penalty function to assess local influence in the possibly heteroskedastic linear regression model with exact restrictions. The restricted maximum likelihood estimators and information matrices are presented for the postulated model. Several perturbation schemes for the local influence method are investigated to identify potentially influential observations. Three real-world examples are included to illustrate and validate our methodology
    Original languageEnglish
    Pages (from-to)227-249
    Number of pages23
    JournalStatistical Methods and Applications
    Volume25
    DOIs
    Publication statusPublished - 2016

    Fingerprint

    Influence Diagnostics
    Local Influence
    Influential Observations
    Linear Model
    Restriction
    Restricted Maximum Likelihood Estimator
    Lagrange multiplier Method
    Information Matrix
    Methodology
    Statistical Modeling
    Penalty Function
    Econometrics
    Linear Regression Model
    Linear Function
    Perturbation
    Model
    Diagnostics

    Cite this

    @article{a958d1305d724ab1bf4dcbf4774ea31a,
    title = "Influence diagnostic analysis in the possibly heteroskedastic linear model with exact restrictions",
    abstract = "The local influence method has proven to be a useful and powerful tool for detecting influential observations on the estimation of model parameters. This method has been widely applied in different studies related to econometric and statistical modelling. We propose a methodology based on the Lagrange multiplier method with a linear penalty function to assess local influence in the possibly heteroskedastic linear regression model with exact restrictions. The restricted maximum likelihood estimators and information matrices are presented for the postulated model. Several perturbation schemes for the local influence method are investigated to identify potentially influential observations. Three real-world examples are included to illustrate and validate our methodology",
    keywords = "information matrix, local influence, restricted least-squares estimator",
    author = "Shuangzhe LIU and Victor Leiva and Tiefeng Ma",
    year = "2016",
    doi = "10.1007/s10260-015-0329-4",
    language = "English",
    volume = "25",
    pages = "227--249",
    journal = "Journal of the Italian Statistical Society",
    issn = "1121-9130",
    publisher = "Physica Verlag",

    }

    Influence diagnostic analysis in the possibly heteroskedastic linear model with exact restrictions. / LIU, Shuangzhe; Leiva, Victor; Ma, Tiefeng.

    In: Statistical Methods and Applications, Vol. 25, 2016, p. 227-249.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Influence diagnostic analysis in the possibly heteroskedastic linear model with exact restrictions

    AU - LIU, Shuangzhe

    AU - Leiva, Victor

    AU - Ma, Tiefeng

    PY - 2016

    Y1 - 2016

    N2 - The local influence method has proven to be a useful and powerful tool for detecting influential observations on the estimation of model parameters. This method has been widely applied in different studies related to econometric and statistical modelling. We propose a methodology based on the Lagrange multiplier method with a linear penalty function to assess local influence in the possibly heteroskedastic linear regression model with exact restrictions. The restricted maximum likelihood estimators and information matrices are presented for the postulated model. Several perturbation schemes for the local influence method are investigated to identify potentially influential observations. Three real-world examples are included to illustrate and validate our methodology

    AB - The local influence method has proven to be a useful and powerful tool for detecting influential observations on the estimation of model parameters. This method has been widely applied in different studies related to econometric and statistical modelling. We propose a methodology based on the Lagrange multiplier method with a linear penalty function to assess local influence in the possibly heteroskedastic linear regression model with exact restrictions. The restricted maximum likelihood estimators and information matrices are presented for the postulated model. Several perturbation schemes for the local influence method are investigated to identify potentially influential observations. Three real-world examples are included to illustrate and validate our methodology

    KW - information matrix

    KW - local influence

    KW - restricted least-squares estimator

    U2 - 10.1007/s10260-015-0329-4

    DO - 10.1007/s10260-015-0329-4

    M3 - Article

    VL - 25

    SP - 227

    EP - 249

    JO - Journal of the Italian Statistical Society

    JF - Journal of the Italian Statistical Society

    SN - 1121-9130

    ER -