Premium
Additive and Non‐additive Genetic Correlations. I. Properties of Sampling Distributions
Author(s) -
Hohls Trevor,
Clarke G. P. Y.
Publication year - 1995
Publication title -
biometrical journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.108
H-Index - 63
eISSN - 1521-4036
pISSN - 0323-3847
DOI - 10.1002/bimj.4710370804
Subject(s) - diallel cross , mathematics , statistics , degrees of freedom (physics and chemistry) , correlation , transformation (genetics) , correlation coefficient , sampling (signal processing) , statistical physics , physics , thermodynamics , chemistry , biology , geometry , biochemistry , botany , hybrid , detector , gene , optics
The sampling distributions of additive and non‐additive genetic correlations estimated from a diallel experiment were determined through a computer simulation study. The properties of the sampling distributions were investigated in terms of the effect of varying the magnitude of the relevant variance components and the true correlation coefficient (Q), as well as the size of the diallel experiment. It was found that there was a high incidence of Q falling outside the range −1 ≦ Q ≦1 when Q was close to 1 or ‐1, and the variance components were small. The degrees of freedom of the correlation components were shown to affect the variability of Q, and it was shown that a diallel experiment, that is to be used for a correlation analysis, should not be smaller than a 12 × 12 half diallel. Use of the Z transformation as a means of testing the significance of Q has been verified in this study and improvements on the degrees of freedom used for the transformation have been discussed.