Parameter estimation in generalized linear models for dependent data

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21340%2F14%3A00222917" target="_blank" >RIV/68407700:21340/14:00222917 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Parameter estimation in generalized linear models for dependent data
Original language description
Generalized linear models represent a class of various statistical models such as linear regression, ANOVA, logistic regression and Poisson regression. These models are widely used in statistical applications and usually assume uncorrelated data. On thecontrary, in longitudinal studies, for example, correlation between observations is encountered. In this case, the method called generalized estimating equations (GEE) can be employed to estimate the parameters of the model. This method gives consistentestimates of the regression parameters of the studied model under mild assumptions about the time dependence. Let b_G denote the estimate computed with GEE considering any given correlation structure and let b_I be the estimate calculated with the Fisherscoring algorithm under the assumption of independence. The issue of efficiency of b_G compared to b_I will be discussed as well as the question of whether it is worth to use the GEE method instead of methods assuming independence regard
Czech name
—
Czech description
—

Publication year
2014
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
Proceedings of Stochastic and Physical Monitoring Systems 2014
ISBN
978-80-01-05616-5
ISSN
—
e-ISSN
—
Number of pages
10
Pages from-to
7-16
Publisher name
ČVUT v Praze
Place of publication
Praha
Event location
Malá Skála
Event date
Jun 23, 2014
Type of event by nationality
EUR - Evropská akce
UT code for WoS article
—

Similar results(10)