You are not currently logged in.
Access JSTOR through your library or other institution:
Restricted Maximum Likelihood (REML) Estimation of Variance Components in the Mixed Model
R. R. Corbeil and S. R. Searle
Vol. 18, No. 1 (Feb., 1976), pp. 31-38
Published by: Taylor & Francis, Ltd. on behalf of American Statistical Association and American Society for Quality
Stable URL: http://www.jstor.org/stable/1267913
Page Count: 8
You can always find the topics here!Topics: Maximum likelihood estimation, Statistical estimation, Statistical variance, Matrices, Estimators, Statistical discrepancies, Variance, Analytical estimating, Mathematical vectors, Estimation methods
Were these topics helpful?See somethings inaccurate? Let us know!
Select the topics that are inaccurate.
Preview not available
The maximum likelihood (ML) procedure of Hartley and Rao  is modified by adapting a transformation from Patterson and Thompson  which partitions the likelihood under normality into two parts, one being free of the fixed effects. Maximizing this part yields what are called restricted maximum likelihood (REML) estimators. As well as retaining the property of invariance under translation that ML estimators have, the REML estimators have the additional property of reducing to the analysis variance (ANOVA) estimators for many, if not all, cases of balanced data (equal subclass numbers). A computing algorithm is developed, adapting a transformation from Hemmerle and Hartley , which reduces computing requirements to dealing with matrices having order equal to the dimension of the parameter space rather than that of the sample space. These same matrices also occur in the asymptotic sampling variances of the estimators.
Technometrics © 1976 American Statistical Association