CCT MENDOZA   20878
CENTRO CIENTIFICO TECNOLOGICO CONICET - MENDOZA
Centro Científico Tecnológico - CCT
congresos y reuniones científicas
Título:
Classification rules for triply multivariate data with an AR(1) correlation structure on the repeated measure over time,
Autor/es:
ANURADHA ROY; RICARDO LEIVA
Lugar:
Atlanta, Georgia, USA..
Reunión:
Conferencia; Spring Meeting of the Eastern North American Region of the International Biometric Society (ENAR 2007 Spring Meeting); 2007
Institución organizadora:
actas del Spring Meeting of the Eastern North American Region of the International Biometric Society
Resumen:
Under the assumption of multivariate normality we study the classification rules for triply multivariate data (multivariate in three directions), where more than one response variable is measured in each experimental unit on more than one site at several time points. It is very common in clinical trial study to collect measurements on more than one variable at different body positions (sites) repeatedly over time. The new classification rules , with and without time effect,  and with certain structured and  unstructured mean vectors and covariance structures, are very efficient in small sample scenario, when the number of observations is not adequate to estimate the unknown variance-covariance matrix. We introduce a parametrically parsimonious model for the classification rules by introducing a equicorrelated (partitioned) covariance matrix on the measurement vector over sites in addition to AR(1) correlation structure on repeated observations over time. Computation algorithms for maximum likelihood estimates of the unknown population parameters are presented. Simulation results show that the introduction of  the sites in the classification rules improves its perfomance.