 |
|
Some contributions to multiple comparison methods for longitudinal data and general linear mixed effects models
Zhengzhi Xia
Paperback. ProQuest, UMI Dissertation Publishing 2011-09-08.
ISBN 9781243732897
|
|
|
|
Hitta bokens lägsta pris
|
Förlagets beskrivning
The assumptions of constant variance and uncorrelated errors that are typically made in general linear models often do not hold for longitudinal data, repeated measurements, and general linear mixed models. The traditional multiple comparison (MCP) methods based on the studentized range distribution and multivariate t distribution are exact only under the assumptions of linearity, normality, constant variance, and uncorrelated errors. The element of MCPs, in many cases, is the simple individual comparison on means of two measurements/treatments. For the two-sample t test, even a moderate correlation among observations will result in serious bias; this problem definitely carries over to multiplicity-adjusted inferences. This thesis briefly reviews the background of various traditional procedures of MCP tests and then investigates some MCPs with longitudinal repeated measures. In these cases the sample errors are inflated/deflated because of the correlation among observations, and results in giving incorrect inferences. In this thesis, we suggest using different ways of dealing with this situation under the first order autoregressive (AR(1)) correlation for various linear mixed models
Liknande böcker
Recensioner
Den här boken har tyvärr inte några recensioner ännu. Om du redan läst boken, skriv en recension!
Recensera boken
Skriv en recension och dela dina åsikter med andra. Försök att fokusera på bokens innehåll. Läs våra instruktioner för mer information.
Some contributions to multiple comparison methods for longitudinal data and general linear mixed effects models
Bokrecensioner » Some contributions to multiple comparison methods for longitudinal data and general linear mixed effects models
|
|
 |
|
|
|
|
