Submitted to: Journal of Dairy Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: March 18, 2012
Publication Date: July 1, 2012
Citation: Cole, J.B., Ehrlich, J.L., Null, D.J. 2012. Short communication: Projecting milk yield using best prediction and the MilkBot lactation model. Journal of Dairy Science. 95(7):4041-4044. Interpretive Summary: Lactation models can be used to compensate for the effect of normal lactation curves on milk production. Three models are evaluated for accuracy and precision by making projections and comparing those projections to observed milk production. Resulting data may be useful in selecting among models and in evaluating the reliability of model-related methods used in research trials or for making management decisions. The nonlinear MilkBot model appears to have an advantage in both accuracy and precision for the data sets tested.
Technical Abstract: The accuracy and precision of three lactation models was estimated by summarizing means and variability in projection error for next-test milk and actual 305-d milk yield (M305) for 50-day intervals in a large DHIA data set. Lactations were grouped by breed (Holstein, Jersey and crossbred) and parity (first versus later). A smaller, single-herd data set with both DHIA data and daily milk weights was used to compare M305 calculated from test-day data with M305 computed by summing daily milk weights. The lactation models tested were best prediction (BP), the nonlinear MilkBot (MB) model, and a null model (NM) based on a stepwise function. Accuracy of models was ranked (best to worst) MB, BP, NM for mature cows and MB, NM, BP for heifers, with MilkBot achieving accuracy in projecting daily milk of 0.5 kg or better in most groups. Models generally showed better accuracy after 50 DIM. Best prediction and NM had low accuracy for crossbred cows and Holstein or Jersey heifers. The MilkBot model appears to be more precise than BP, and NM had low precision, especially for M305. Regression of model-generated M305 on summed M305 showed BP and MB to be equally efficient in ranking lactations, but MB was better at quantifying differences.