Predicting growth and performance of Holstein steers

Refinements and additions were made to the growing cattle portion of the model of Fox et al. (1988) for predicting growth of Holstein steers. Based on observations of animal weight, DM intake (DMI), ADG, feed per gain (FPG) and diet ME, the modified model was tested on a database that included 299 feeding periods with Holstein steers fed in research trials over a period of 15 yr at Cornell University, the University of Minnesota and the University of Wisconsin. The modified model accounted for 93, 56 and 68% of the variation in DM intake, daily gain and feed required per unit gain, respectively. Averaged over all 299 periods in the database, the error in predicting DM intake, daily gain and feed required per unit gain was 1%, 4% and 2%, respectively. Analysis of residuals indicated that ME level did not influence the difference between predicted and observed values for DMI, ADG and FPG. Residuals analysis indicated that BW did not influence the difference between predicted and observed values for ADG and FPG but DMI of heavier cattle was overpredicted (P less than .01). These results indicated that the 8% increase in DMI for cattle over 318 kg was not warranted for the Holstein steers in this database, which were predominantly less than 15 mo at slaughter. The results of the validation indicate that this model accurately predicted performance of Holstein steers.