Wiggans G R, VanRaden P M, Powell R L
Animal Improvement Programs Laboratory, Agricultural Research Service, Beltsville, MD 20705-2350.
J Dairy Sci. 1992 Oct;75(10):2834-9. doi: 10.3168/jds.S0022-0302(92)78046-1.
Canadian and US genetic evaluations for July 1991 were combined for 3304 Holstein bulls evaluated in Canada that had US progeny or a cross-reference code. Canadian evaluations of the bulls and their sires and dams were converted to US PTA. Combined PTA were weighted averages of progeny information from both countries and parent average. Parent average was recomputed from the sire's combined evaluation and the dam's evaluation with the most daughter equivalents. Bulls were processed in birth year order so that combined evaluation of sire was available. Progeny contribution was adjusted to remove the influence of the bull's evaluation on progeny evaluations through their parent averages, which left only the portion that was due to progeny records and grandprogeny information. A weighted average of adjusted progeny contributions was combined with parent average to form a combined PTA more accurately than possible by averaging domestic and converted PTA. Combined reliability was computed by summing daughter equivalents from progeny and from updated evaluations of parents. Canadian daughter equivalents were multiplied by .9 to approximate US values. Combined evaluations simplify bull selection by providing comparable and complete information for bulls evaluated in two countries.
1991年7月加拿大和美国的遗传评估数据被合并，涉及在加拿大接受评估的3304头拥有美国后代或交叉参考代码的荷斯坦公牛。这些公牛及其父系和母系的加拿大评估数据被转换为美国的预测传递能力（PTA）。合并后的PTA是来自两国后代信息和父母平均值的加权平均数。父母平均值是根据父亲的合并评估以及拥有最多女儿等效值的母亲的评估重新计算得出的。公牛按照出生年份顺序进行处理，以便获得父亲的合并评估。通过父母平均值调整后代贡献，以消除公牛评估对其后代评估的影响，这样就只剩下了由后代记录和孙代信息所导致的部分。调整后的后代贡献的加权平均数与父母平均值相结合，从而形成一个比直接平均国内和转换后的PTA更准确的合并PTA。合并可靠性是通过将后代以及父母更新评估中的女儿等效值相加来计算的。加拿大的女儿等效值乘以0.9以近似美国的值。合并评估通过为在两个国家评估的公牛提供可比且完整的信息，简化了公牛的选择过程。
