Skip to main content
ARS Home » Research » Publications at this Location » Publication #125896

Title: HETEROGENEITY IN (C0)VARIANCE STRUCTURES OF TEST-DAY YIELDS

Author
item GENGLER, N - GEMBLOUX AGRIC UNIV
item Wiggans, George

Submitted to: Interbull Annual Meeting Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 8/30/2001
Publication Date: N/A
Citation: N/A

Interpretive Summary: A method to account for differences in variation across herd and stage of lactation was investigated. Data were first-lactation test-day milk, fat and protein yields from New York, Wisconsin, and California herds that were adjusted additively for age and lactation stage. A random regression model was applied to estimate the genetic effects. Phenotypic variances were modeled using a mixed model. Heritabilities and permanent environment showed the expected pattern. Herd, time period effects explained some of the phenotypic variance differences especially at the beginning of the lactation. Variances increased with time, size of herd and milking frequency. Month of test had only a small influence. Low and high milk production level showed increased variances, as did late and especially early lactation stages. These results show that variances are heterogeneous, and demonstrates a method of accounting for this heterogeneity simultaneously with estimation of genetic values.

Technical Abstract: First-lactation test-day milk, fat and protein yields from New York, Wisconsin, and California herds were adjusted additively for age and lactation stage. A random regression model with third-order Legendre polynomials for permanent environmental and genetic effects was used. This model also included a random effect with the same polynomial regressions for 2-yr intervals within herd (herd, time period of calving effect). Phenotypic variances were modeled using a mixed model. Heritabilities and permanent environment showed the expected pattern. Herd, time period effects explained some of the phenotypic variance differences especially at the beginning (12% to 20%) of the lactation. Variances increased with time, size of subclass and milking frequency. Month of test had only a very limited influence. Low and high milk production level showed increased variances, as did late and especially early lactation stages. Repeatabilities of variances observed for a given herd, test-day, frequency class across nested variance subclasses based on lactation stage were 14% to 17%.