Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Animal Genomics and Improvement Laboratory » Research » Publications at this Location » Publication #413685
Research Project: Increasing Accuracy of Genomic Prediction, Developing Algorithms, Selecting Markers, and Evaluating New Traits to Improve Dairy Cattle

Location: Animal Genomics and Improvement Laboratory

Title: Variance reduction and measurement errors in estimating lactation milk yields using Best Prediction: An analytical review

Author
item WU, XIAO-LIN - Council On Dairy Cattle Breeding
item Van Raden, Paul
item COLE, JOHN - Council On Dairy Cattle Breeding
item NORMAN, HOWARD - Council On Dairy Cattle Breeding

Submitted to: Journal of Dairy Science Communications
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/17/2024
Publication Date: 3/1/2025
Citation: Wu, X., Van Raden, P.M., Cole, J.B., Norman, H.D. 2025. Variance reduction and measurement errors in estimating lactation milk yields using best prediction: An analytical review. Journal of Dairy Science Communications. 6(2):231–236. https://doi.org/10.3168/jdsc.2024-0622.
DOI: https://doi.org/10.3168/jdsc.2024-0622

Interpretive Summary: Since 1999, Best Prediction has been utilized in the United States to estimate unobserved daily and lactation yields using known test-day yields. This method has proven more accurate than its predecessors in estimating lactation yields but tends to reduce the variance of estimated yields compared to actual yields, and measurement errors can occur due to inaccuracies in milk yield measurements. This paper provides an analytical review of Best Prediction, focusing on issues related to variance reduction and measurement errors, and presents illustrative examples. Precisely adjusting projected lactation yields to account for measurement errors remains challenging, and alternative strategies may deserve further investigation.

Technical Abstract: Since 1999, Best Prediction has been utilized in the United States to estimate unobserved daily and lactation yields from known test-day yields. This method has proven more accurate than its predecessors. However, it presents two significant challenges in practice. Firstly, Best Prediction tends to reduce the variance of estimated yields compared to actual yields, which is problematic for genetic evaluations as it can significantly underestimate genetic variations. Secondly, measurement errors can occur when projecting lactation yields from incomplete or inaccurate test-day records. These errors can propagate through the calculation process, adversely affecting the accuracy of lactation yield estimations and the subsequent genetic evaluations. This paper provides an analytical review of Best Prediction, with a particular focus on issues related to variance reduction and measurement errors. We demonstrate how measurement errors can be intrinct Best Prediction and propose potential mitigation strategies. Illustrative examples are presented and discussed, highlighting the practical challenges and potential solutions for managing measurement errors effectively. While precisely adjusting projected lactation yields to account for measurement errors remains challenging, alternative strategies such as error-in-variance models offer that promises avenues for improving estimation accuracy are worth further investigating.