Premium
Multivariate one‐sided tests for nonlinear mixed‐effects models
Author(s) -
Wang Tao,
Wu Lang
Publication year - 2013
Publication title -
canadian journal of statistics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.804
H-Index - 51
eISSN - 1708-945X
pISSN - 0319-5724
DOI - 10.1002/cjs.11180
Subject(s) - multivariate statistics , multivariate analysis , wald test , contingency table , multivariate analysis of variance , statistics , econometrics , likelihood ratio test , statistical hypothesis testing , score test , test (biology) , multivariate normal distribution , computer science , mathematics , geology , paleontology
Multivariate one‐sided hypotheses testing problems arise frequently in practice. Various tests have been developed, but mainly focused on multivariate normal distributions or contingency tables. In this article, motivated by HIV viral dynamic models, we study multivariate one‐sided tests in nonlinear mixed‐effects (NLME) models, for which no previous research has been carried out. We propose a likelihood ratio test (LRT), a Wald test, and a score test. Theoretical results are presented and computational issues are discussed. The proposed methods are evaluated using simulations. A real data example is presented to illustrate the methods. The Canadian Journal of Statistics 41: 453–465; 2013 © 2013 Statistical Society of Canada