Dm. Zucker et al., Improved small sample inference in the mixed linear model: Bartlett correction and adjusted likelihood, J ROY STA B, 62, 2000, pp. 827-838
Citations number
24
Categorie Soggetti
Mathematics
Journal title
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY
The mixed linear model is a popular method for analysing unbalanced repeate
d measurement data. The classical statistical tests for parameters in this
model are based on asymptotic theory that is unreliable in the small sample
s that are often encountered in practice. For testing a given fixed effect
parameter with a small sample, we develop and investigate refined likelihoo
d ratio (LR) tests. The refinements considered are the Bartlett correction
and use of the Cox-Reid adjusted likelihood; these are examined separately
and in combination. We illustrate the various LR tests on an actual data se
t and compare them in two simulation studies. The conventional LR test yiel
ds type I error rates that are higher than nominal. The adjusted LR test yi
elds rates that are lower than nominal, with absolute accuracy similar to t
hat of the conventional LR test in the first simulation study and better in
the second. The Bartlett correction substantially improves the accuracy of
the type I error rates with either the conventional or the adjusted LR tes
t. In many cases, error rates that are very close to nominal are achieved w
ith the refined methods.