Regression Factors for Predicting a Succeeding Complete Lactation Milk Record from Part Lactation Records

Open Access
Publisher Website
Google Scholar
Abstract
The age- and season-corrected monthly D.H.I.A. milk records of 9036 cows in 374 New York herds and the succeeding record of 2358 of. these cows in 226 herds are used to develop regression factors for predicting a succeeding complete lactation milk yield from part lactation records. A single test day record from the 4th, 5th or 6th month of a lactation has a correlation of 0.50 with the succeeding complete lactation. This correlation is nearly as large as the correlation between 2 successive complete lactation yields which is 0.55. A linear function of the 1st 5 monthly test records predicts a complete 2nd record as accurately as the total yield of the 1st lactation. Linear functions for the 1st 7 monthly records are even more accurate. Prediction equations utilizing bimonthly test records are more highly correlated (0.57) with 2nd complete records than complete 1st records and are nearly as accurate as linear functions of 10 monthly test day records.