GOODNESS-OF-FIT TESTS FOR GEE MODEL | METHODS AND APPLICATIONS

Md. Abdus Salam Akanda (u. a.)

ISBN 10: 3639351223 / ISBN 13: 9783639351224

Edité par VDM Verlag Dr. Müller, 2011

GOODNESS-OF-FIT TESTS FOR GEE MODEL | METHODS AND APPLICATIONS | Md. Abdus Salam Akanda (u. a.) | Taschenbuch | Einband - flex.(Paperback) | Englisch | 2011 | VDM Verlag Dr. Müller | EAN 9783639351224 | Verantwortliche Person für die EU: OmniScriptum GmbH & Co. KG, Bahnhofstr. 28, 66111 Saarbrücken, info[at]akademikerverlag[dot]de | Anbieter: preigu. N° de réf. du vendeur 106955321

Synopsis :

During the last decades, a vertical exploration of statistical methodology in the ground of longitudinal data has been observed. One of the most exciting developments in this area is the Generalized Estimating Equation (GEE). This is a widely used approach that does not require the complete specification of the joint distribution of repeated measurements. In addition, GEE takes into account which results in attaining more efficiency in estimating parameters of marginal models. The model used in this study was developed by Barnhart & Williamson (1998). To test the goodness-of-fit we have applied model based and empirically corrected tests according to their implication. In their suggested GEE approach the correlation between two responses was not considered. Here an alternative procedure is proposed based on GEE where the correlation between two responses was considered. This study has also indicated that the model with only main effects did not fit the data well. There is a significant region, time and interaction effect. The identity correlation structure provides most efficient estimates to represent the relationship among the covariates and the responses.

Détails bibliographiques

Titre : GOODNESS-OF-FIT TESTS FOR GEE MODEL | ...
Éditeur : VDM Verlag Dr. Müller
Date d'édition : 2011
Reliure : Taschenbuch
Etat : Neu