| Conditional and exact tests in crossover trials. | |
| | |
MedLine Citation:
|
PMID: 10709805 Owner: NLM Status: MEDLINE |
Abstract/OtherAbstract:
|
Generalized linear models are developed for crossover trials with no carryover effects and fixed subject effects. A general multinominal model for the distribution of data is considered. This subsumes both binary and categorical data. Conditional inferences eliminate subject effects by conditioning on their sufficient statistics. For normal data, least-squares analysis is exact with identical treatment inferences from unconditional and conditional analyses. For Poisson data, unconditional and conditional analyses are also identical, but for multinomial data this is not the case and the unconditional analysis is invalid. For multinomial data, asymptotic tests of both treatment effects and goodness of fit are unreliable with small samples. Procedures for exact tests are developed to overcome such problems, using enumeration, random sampling, and a hybrid of importance sampling and enumeration. A four-period binary crossover trial is used to illustrate an exact test of treatment effects by a two-stage sampling procedure based on a factorization of the conditional distribution of the sufficient statistics. An exact test of goodness of fit on the same data illustrates a two-stage scheme mixing importance sampling and enumeration. |
| | |
Authors:
|
M Patefield |
Related Documents
:
|
270465 - The clinical trial in multidisciplinary research. 10790975 - Testing instance models of face repetition priming. 20840525 - A new tool for the validation of umbilical cord acid-base data. 15606425 - Evaluation of community-intervention trials via generalized linear mixed models. 21316685 - Prediction of the physicochemical properties of gasoline by comprehensive two-dimension... 15072605 - Biochemical recurrence and survival prediction models for the management of clinically ... |
Publication Detail:
|
Type: Journal Article |
Journal Detail:
|
Title: Journal of biopharmaceutical statistics Volume: 10 ISSN: 1054-3406 ISO Abbreviation: J Biopharm Stat Publication Date: 2000 Feb |
Date Detail:
|
Created Date: 2000-04-17 Completed Date: 2000-04-17 Revised Date: 2007-11-15 |
Medline Journal Info:
|
Nlm Unique ID: 9200436 Medline TA: J Biopharm Stat Country: UNITED STATES |
Other Details:
|
Languages: eng Pagination: 109-29 Citation Subset: IM |
Affiliation:
|
Department of Applied Statistics, The University of Reading, United Kingdom. |
Export Citation:
|
APA/MLA Format Download EndNote Download BibTex |
| MeSH Terms | |
Descriptor/Qualifier:
|
Binomial Distribution Cross-Over Studies Double-Blind Method Humans Likelihood Functions Multivariate Analysis Poisson Distribution Randomized Controlled Trials as Topic / methods* Sampling Studies Statistics as Topic / methods* |
From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine
Previous Document: A comparison of urn designs for randomized clinical trials of K > 2 treatments.
Next Document: Neurophysiologic basis of functional neuroimaging: animal studies.