Open AccessCOMPARISON OF SINGLE AND MULTIPLE TRAIT BEEF SIRE EVALUATIONS
Author(s) -
L.R. Schaeffer,
J. W. Wilton
Publication year - 1981
Publication title -
canadian journal of animal science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.377
H-Index - 64
eISSN - 1918-1825
pISSN - 0008-3984
DOI - 10.4141/cjas81-068
Subject(s) - trait , sire , selection (genetic algorithm) , ice calving , biology , statistics , mathematics , zoology , genetics , pregnancy , computer science , lactation , artificial intelligence , programming language
Beef bulls of 13 breeds were evaluated from progeny records for calving ease, weaning weight and yearling weight using single and multiple trait best linear unbiased prediction techniques. Multiple trait procedures allowed more bulls to be evaluated for all traits and gave slightly greater variances to sire evaluations than the single trait method. The correlations of proofs between single and multiple trait methods were above 0.9 for the majority of the breeds for each trait, with the lowest correlations being between yearling weight proofs from the two methods. The correlations of proofs between traits within the multiple trait analysis were greater in absolute magnitude than within the single trait analyses. Multiple-trait procedures seem to be advantageous for lowering the standard errors of prediction for young bulls with their first proof, and for removing the effects of selection for weaning weight from the yearling weight proofs.