Extending Part Lactation Milk Records by Regression Ignoring Herd Effects

Authors:
  • LVL.D. Van Vleck
    L.D. Van Vleck
Open Access
Publisher website
Google Scholar
Abstract
SuD/[ 1vf ARY Regression coefficients for predicting 10-mo. milk yield from single test-day records, from cumulative test-day records of less than 10 too. duration, from bimonthly and trimonthly test-day records, and from other combinations of test-day records are estimated ignoring herd effects. The records, adjusted for age a~ calving and season of calving, used for estimating these coefficients were from 9,036 Holstein cows in five New York counties. These regression coefficients are compared with intraherd coefficients reported earlier which were estimated from the same data. The relative efficiencies of the two procedures are compared to determine which procedure is more practical. For most combinations of records regression ignoring herd effects is less than 20% less efficient and in many cases less than 10% less efficient than regression considering herd effects. It is concluded that prediction of 10-too. milk yield by regression ignoring herd effects is more practical for most situations, because of computational advantages although the accuracy of prediction is slightly less in all situations than for intra-herd prediction.

This publication has 10 references indexed in Scilit: