Submitted to: Interbull Annual Meeting Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 8/28/2003
Publication Date: 10/1/2003
Citation: Powell, R.L., Sanders, A.H., Norman, H.D. 2003. Accuracy of foreign dairy bull evaluations in predicting US evaluations for yield. International Bull Evaluation Service Bulletin 31:171-174.
Interpretive Summary: Use of US dairy genetics by other countries resulted in their upgraded populations of Holsteins being of interest as a source of bull semen for US dairies. This project investigated the accuracy of genetic estimates of merit for yield based on foreign daughters as predictors US performance. Foreign evaluations had been processed by the International Bull Evaluation Service (Interbull) with the objective of placing those evaluations on a scale directly comparable to that of evaluations from US data. That hypothesis was tested by comparing the most recent Interbull evaluation that did not include US daughters with the August US evaluation from only US daughters. Both sets of evaluations were required to be of reasonably high accuracy such that only 281 bulls qualified. Average differences were near zero, demonstrating that the foreign data gave a generally unbiased prediction of the US evaluation although the variation showed that the estimated genetic transmitting ability for about one-third of the bulls differed by more than 167 kg of milk. Correlations for milk, fat, and protein evaluations were .87 to .90 compared to an expected correlation of .89. About two-thirds (67%) of the bulls were from Canada, 15% from The Netherlands, 5% each from France and Germany, and the remaining bulls from 7 other countries. The results for the Canadian bulls were compared with those from the other countries and no important differences were found. The small average differences between evaluations from foreign and US daughters and the high correlations indicate that Interbull evaluations based solely on foreign daughters are good predictors of the US evaluations for yield.
Technical Abstract: The addition of foreign daughter data to domestic dairy bull genetic evaluations has been shown to improve prediction of future domestic evaluations in a study of mainly US bulls. This study focused on foreign bulls by evaluating the accuracy of Interbull evaluations (foreign daughters) in predicting the latest US yield evaluations (US daughters). The most recent February or August Interbull evaluations without US daughters from 1995 through 2003 were matched with August 2003 USDA evaluations based only on US daughters for the 281 Holstein bulls having reliability of at least 80%. This provided pairs of evaluations based on different daughters (foreign or US). For the Interbull evaluations, mean reliability on the US scale was 88% and the mean US reliability for August 2003 evaluations was 91%. Correlations between these Interbull and US evaluations were .90, .87, and .90. for milk, fat, and protein, respectively. Expected correlations were .89. Interbull evaluations averaged nearly the same as the UDSA evaluations based only on US daughters. About two-thirds (67%) of the bulls were from Canada, 15% from The Netherlands, 5% each from France and Germany, and the remaining bulls from 7 other countries. Differences between the Interbull and USDA evaluations were near zero for both groups of bulls (Canadian and other) and the standard deviations of those differences were nearly identical. Because the foreign bulls had graduated from progeny test programs before becoming available here, there had been selection for positive Mendelian sampling, causing parent average to be a poor underestimate of merit. The small average differences between evaluations from foreign and US daughters and the high correlations indicate that Interbull evaluations based solely on foreign daughters are good predictors of the US evaluations for yield.