Ανακοινώνεται ότι την Τετάρτη 24 Απριλίου 2002, και ώρα 13:00-14:00, θα δοθούν ομιλίες από μέλη ΔΕΠ του Τμήματός μας, στην αίθουσα Α32 του κτιρίου Αντωνιάδου, σύμφωνα με το παρακάτω πρόγραμμα.

 Τετάρτη, 24 Απριλίου 2002
 Βασδέκης Βασίλειος, Λέκτορας του Τμήματος Στατιστικής.

 Τίτλος σεμιναρίου: "Covariance Adjustment Methods in the Analysis of Repeated Measures Data"


 Βασδέκης Βασίλειος
 Covariance Adjustment Methods in the Analysis of Repeated Measures Data

 ABSTRACT
 The GMANOVA model is considered when information for one characteristic is obtained from n individuals. It is assumed that measurements are made at q  time points. The covariance adjusted estimator is the OLS estimator adjusted using analysis of covariance. The covariates are artificially  constructed and are obtained from the space vertical to that derived from the design matrix. The form of the estimator is obtained as a GLS estimator using a weight matrix of reduced rank unless all available covariates are > used. The method has similarities to REML with the difference that the  latter is a likelihood based approach. The choice of the appropriate  covariate combination is faced by considering AIC and cross validation. A comparison between the two methods is accomplished through a small  simulation study.

(c) 2001 Ioannis Ntzoufras

Last revised: 26-10-2001.