Goodness-of-fit test in parametric mixed effects models based on estimation of the error distribution

We address the problem of testing for a parametric function of fixed effects in mixed models. We propose a test based on the distance between two empirical error distribution functions, which are constructed from residuals calculated under the opposing hypotheses. The proposed test statistic has power against all alternatives, and its asymptotic distribution is derived. A simulation study shows that the test outperforms others in the literature. The test is applied to longitudinal data from an AIDS clinical trial and a growth study.

• metadata
• relations

Document type	Article de périodique (Journal article) – Article de recherche
Access type	Accès interdit
Publication date	2016
Language	Anglais
Journal information	"Biometrika" - Vol. 103, no. 1, p. 133-146 (2016)
Peer reviewed	yes
Publisher	Oxford University Press ((United Kingdom) Oxford)
issn	0006-3444
e-issn	1464-3510
Publication status	Publié
Affiliation	UCL - SSH/LIDAM/ISBA - Institut de Statistique, Biostatistique et Sciences Actuarielles
Keywords	Bootstrap ; Empirical distribution ; Local polynomial estimation ; Mixed model ; Residual
Links	http://biomet.oxfordjournals.org/content/early/2016/02/01/biomet.asv061.refs?sid=18411786-6908-4862-8df9-b97681e135ac http://hdl.handle.net/2078.1/171515[Handle] https://doi.org/10.1093/biomet/asv061[DOI]

See also
	[boreal:128822]	Goodness-of-fit Test in Parametric Mixed-Effects Models based on the Estimation of the Error Distribution