Listy Biometryczne - Biometrical Letters, Vol. 40 (2003), No. 1, 1- 7
|
ESTIMATING AND TESTING OF VARIANCE COMPONENTS: AN APPLICATION TO A GRAPEVINE EXPERIMENT
Miguel Fonseca1, Joao Tiago Mexia1 and Roman Zmyslony2 1Department of Mathematics, Faculty of Science and Technology, New University of Lisbon, Monte da Caparica 2829-516, Caparica, Portugal2Institute of Mathematics, University of Zielona Góra, Podgórna 50, 65-246 Zielona Góra, Poland |
The problem of genetic homogeneity for grapevines castes is of great practical interest. An experiment to assess such homogeneity was analysed. The full discussion of the results of that experiment required obtaining uniformly minimum variance unbiased estimators (UMVUE) for all variance components of a three factor random design and showing how to test all of them. This problem was solved using generalized F statistics. They are quotients of convex combinations of independent chi-squares divided by their degrees of freedom.
variance components, uniformly minimum variance unbiased estimators (UMVUE), linear nested random model, genetic homogeneity, grapevine.
