Monments.Standard deviations (Table 1) were lowest for thetwo BP estimates for all testing environments be-cause those methods regressed outlier test day yieldseither toward breed mean or herd mean. Each cor-responding standard deviation was greater for APthan for MN testing except for the BP method con-sidering herd mean. The large variation is not unex-pected considering that no milking times wererecorded in the Canadian data set and that daily yieldfor AP testing was assumed to be twice the single-milking weight. For TRI testing, the standard devia-tion for TIM was considerably larger than those fromMN and AP testing; however, the standard deviationsfor both BP methods were smaller than that for MNtesting. The standard deviation of estimated yieldfrom TRI testing was smaller than that from AP
Comparison of Test Interval and Best Prediction Methods for Estimation of Lactation Yield from Monthly, a.m.–p.m., and Trimonthly Testing (PDF Download Available). Available from: https://www.researchgate.net/publication/13221765_Comparison_of_Test_Interval_and_Best_Prediction_Methods_for_Estimation_of_Lactation_Yield_from_Monthly_am-pm_and_Trimonthly_Testing [accessed May 8, 2016].