Evaluating Relationships among Immature Measures of Size, Shape and Performance of Beef Bulls IV. Regression Models for Predicting Postweaning Performance of Young Hereford and Angus Bulls Using Preweaning Measures of Size and Shape

Abstract

Ten skeletal measures and the principal components from an analysis of these 10 measures were obtained on 550 individually fed Hereford (H) and Angus (A) bulls at 4 and again at 8 months of age. The skeletal measures and principal components were used separately in stepdown regression models to predict postweaning gain, feed conversion, feed consumption and final test weight. The coefficients of multiple determination indicated approximately 25% of the variation in test gain could be explained from combinations of skeletal measures and only about 16% from combinations of the principal components, i.e., size and shape. Nearly 65% of the variation in final test weight was accounted for by variation in preweaning body measures. Approximately 45% of the variation in feed consumption was explained using preweaning body measures. With a measure of weight on test, a predicted final weight and predicted feed consumption, the animal's gain on test and feed conversion can be estimated. The accuracy with which feed consumption and final test weight were predicted may be sufficient to justify careful consideration of projecting bull performance at later ages from early test performance or supplemented preweaning performance. The principal components were used to measure differences in overall size and shape. As obtained in this study, they are linear functions of the 10 skeletal measures and, as such, would be expected to be nearly as effective as the individual body measures in predicting performance. However, the results imply that preweaning shapes, measured objectively, were related to subsequent performance; and, if evaluated independently from other body measures, should improve the predictability of postweaning performance. The various prediction models for the two breed groups retained several body measures that were common to both groups; but length was the most frequently omitted measure in immature H bulls, and width measures were absent in nearly all prediction equations of A bulls. The models for each of the four postweaning traits were different; and measures of height, width, length, heart girth and chest depth, in addition to weight, would be necessary in any model for general application.