BIROn - Birkbeck Institutional Research Online

    Modelling covariance structure in bivariate marginal models for longitudinal data

    Xu, Jing (2012) Modelling covariance structure in bivariate marginal models for longitudinal data. Biometrika 2012 (3), pp. 649-662. ISSN 1464-3510.

    [img] Text (Refereed)
    649.full.pdf - Published Version of Record
    Restricted to Repository staff only

    Download (205kB)


    It can be more challenging to efficiently model the covariance matrices for multivariate longitudinal data than for the univariate case, due to the correlations arising between multiple responses. The positive-definiteness constraint and the high dimensionality are further obstacles in covariance modelling. In this paper, we develop a data-based method by which the parameters in the covariancematrices are replaced by unconstrained and interpretable parameterswith reduced dimensions. The maximum likelihood estimators for the mean and covariance parameters are shown to be consistent and asymptotically normally distributed. Simulations and real data analysis show that the new approach performs very well even when modelling bivariate nonstationary dependence structures.


    Item Type: Article
    Keyword(s) / Subject(s): bivariate marginal model, block triangular factorization, covariance modelling, log-innovation matrix modelling, longitudinal data, matrix logarithm
    School: School of Business, Economics & Informatics > Economics, Mathematics and Statistics
    Depositing User: Jing Xu
    Date Deposited: 25 Jan 2013 10:04
    Last Modified: 17 Jun 2021 04:17


    Activity Overview
    6 month trend
    6 month trend

    Additional statistics are available via IRStats2.

    Archive Staff Only (login required)

    Edit/View Item Edit/View Item