Kenward, M.G. and Roger, J.H. (1997) Small sample inference for fixed effects from restricted maximum likelihood. Biometrics, 53 (3). pp. 983-997. ISSN 0006-341X.
|The full text of this publication is not available from this repository. (Contact us about this Publication)|
Restricted maximum likelihood (REML) is now well established as a method for estimating the parameters of the general Gaussian linear model with a structured covariance matrix, in particular for mixed linear models. Conventionally, estimates of precision and inference for fixed effects are based on their asymptotic distribution, which is known to be inadequate for some small-sample problems. In this paper, we present a scaled Wald statistic, together with an F approximation to its sampling distribution, that is shown to perform well in a range of small sample settings. The statistic uses an adjusted estimator of the covariance matrix that has reduced small sample bias. This approach has the advantage that it reproduces both the statistics and F distributions in those settings where the latter is exact, namely for Hotelling T-2 type statistics and for analysis of variance F-ratios. The performance of the modified statistics is assessed through simulation studies of four different REML analyses and the methods are illustrated using three examples.
|Uncontrolled keywords:||alpha design; ante-dependence; crossover trial; mixed models; residual maximum likelihood; small sample approximation|
|Subjects:||Q Science > QA Mathematics (inc Computing science)
Q Science > QH Natural history > QH301 Biology
|Divisions:||Faculties > Science Technology and Medical Studies > School of Biosciences|
|Depositing User:||M.A. Ziai|
|Date Deposited:||21 Apr 2009 02:57|
|Last Modified:||21 Apr 2009 02:57|
|Resource URI:||http://kar.kent.ac.uk/id/eprint/18141 (The current URI for this page, for reference purposes)|
- Depositors only (login required):