%0 Journal Article %1 pourahmadi1999joint %A Pourahmadi, M %D 1999 %J Biometrika %K Cholesky_decomposition covariance_estimation covariance_matrix statistics %N 3 %P 677-690 %R 10.1093/biomet/86.3.677 %T Joint mean-covariance models with applications to longitudinal data: unconstrained parameterisation %U http://biomet.oxfordjournals.org/content/86/3/677.abstract %V 86 %X We provide unconstrained parameterisation for and model a covariance using covariates. The Cholesky decomposition of the inverse of a covariance matrix is used to associate a unique unit lower triangular and a unique diagonal matrix with each covariance matrix. The entries of the lower triangular and the log of the diagonal matrix are unconstrained and have meaning as regression coefficients and prediction variances when regressing a measurement on its predecessors. An extended generalised linear model is introduced for joint modelling of the vectors of predictors for the mean and covariance subsuming the joint modelling strategy for mean and variance heterogeneity, Gabriel's antedependence models, Dempster's covariance selection models and the class of graphical models. The likelihood function and maximum likelihood estimators of the covariance and the mean parameters are studied when the observations are normally distributed. Applications to modelling nonstationary dependence structures and multivariate data are discussed and illustrated using real data. A graphical method, similar to that based on the correlogram in time series, is developed and used to identify parametric models for nonstationary covariances.

@article{pourahmadi1999joint, abstract = {We provide unconstrained parameterisation for and model a covariance using covariates. The Cholesky decomposition of the inverse of a covariance matrix is used to associate a unique unit lower triangular and a unique diagonal matrix with each covariance matrix. The entries of the lower triangular and the log of the diagonal matrix are unconstrained and have meaning as regression coefficients and prediction variances when regressing a measurement on its predecessors. An extended generalised linear model is introduced for joint modelling of the vectors of predictors for the mean and covariance subsuming the joint modelling strategy for mean and variance heterogeneity, Gabriel's antedependence models, Dempster's covariance selection models and the class of graphical models. The likelihood function and maximum likelihood estimators of the covariance and the mean parameters are studied when the observations are normally distributed. Applications to modelling nonstationary dependence structures and multivariate data are discussed and illustrated using real data. A graphical method, similar to that based on the correlogram in time series, is developed and used to identify parametric models for nonstationary covariances.}, added-at = {2013-08-20T00:40:35.000+0200}, author = {Pourahmadi, M}, biburl = {https://www.bibsonomy.org/bibtex/26fa5137539f7031c2c46ae7c19c6ee3d/peter.ralph}, doi = {10.1093/biomet/86.3.677}, eprint = {http://biomet.oxfordjournals.org/content/86/3/677.full.pdf+html}, interhash = {771a974b652327944abecb03b7a9be8f}, intrahash = {6fa5137539f7031c2c46ae7c19c6ee3d}, journal = {Biometrika}, keywords = {Cholesky_decomposition covariance_estimation covariance_matrix statistics}, number = 3, pages = {677-690}, timestamp = {2013-08-20T00:40:36.000+0200}, title = {Joint mean-covariance models with applications to longitudinal data: unconstrained parameterisation}, url = {http://biomet.oxfordjournals.org/content/86/3/677.abstract}, volume = 86, year = 1999 }