Accuracy of Monthly, Bimonthly, and Trimonthly Dairy Herd Improvement Association Records

Abstract

The DHIA centering date method of testing, used in the United States, the DHIA method modified to adjust production in the first and last periods, according to the day of lactation when tested, and the DHIA method modified to adjust production for the day of lactation when first tested were simulated on daily milk weights of 369 Holstein lactations from the Cornell University herd. The DHIA method of testing over-estimated actual production by an average of 57.3, 80.8, and 80.8 kg of milk for monthly, bimonthly, and trimonthly testing, respectively. Adjusting production in the first and last test periods reduced the mean deviation from actual production to -0.5, 1.5, and -2.9 kg of milk for monthly, bimonthly, and trimonthly DHIA testing. The variance of the deviation from actual production was reduced to 90.7, 89.0, and 75.2% of the variance of deviations in the unadjusted centering-day method for monthly, bimonthly, and trimonthly DHIA testing, respectively. Adjusting production in only the first test period yielded mean deviations from actual production of 1.4, 11.2, and 20.3 kg of milk, and reduced the variance of the deviations to 91.8, 95.1, and 84.4% of the variance of deviations of the unadjusted DHIA method from actual production for monthly, bimonthly, and trimonthly, respectively. It was concluded that correcting test-day production for the stage of lactation in the first and last test periods merits consideration, because of the more accurate estimate of actual production obtained and the reduction in the variance of the deviations of DHIA estimates from actual yield. © 1968, American Dairy Science Association. All rights reserved.