Author/Authors
KAYA BAŞAR, Ebru Akdeniz University - Faculty of Agriculture - Department of Animal Science, Biometry and Genetics Unit, Turkey , FIRAT, Mehmet Ziya Akdeniz University - Faculty of Agriculture - Department of Animal Science, Biometry and Genetics Unit, Turkey
Title Of Article
COMPARISON OF METHODS OF ESTIMATING VARIANCE COMPONENTS IN BALANCED TWO-WAY RANDOM NESTED DESIGNS
شماره ركورد
41875
Abstract
The objective of this paper is to compare different methods of estimating variance components, such as analysis of variance (ANOVA), maximum likelihood (ML), restricted maximum likelihood (REML) and Bayesian. The ANOVA method for estimating variance components is based on equating the mean squares to their expectations. However, the problem with this method is that it gives negative estimates of variance components. This can be overcame by the use of likelihood based methods and Bayesian. In this study, four different methods of estimating variance components were compared and also demonstrated how these methods overcome the problem of negative estimates of variance components in balanced two-way random nested designs.
From Page
1
NaturalLanguageKeyword
Variance components , Bayesian analysis , REML , ML.
JournalTitle
Eskişehir Technical University Journal of Science and Technology B - Theoretical Sciences
To Page
10
JournalTitle
Eskişehir Technical University Journal of Science and Technology B - Theoretical Sciences
Link To Document