Skip to main content
ARS Home » Research » Publications at this Location » Publication #99180
Title: STRATEGIES FOR COMBINING TEST DAY EVALUATIONS INTO AN INDEX FOR LACTATION PERFORMANCE

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

Submitted to: American Dairy Science Association Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 6/20/1999
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: Test day models generate many genetic values that reflect different aspects of lactation performance. Objectives of an index combining theses values is to create an optimum selection criterion. Test day models are displacing lactation models as the preferred model for genetic evaluation of dairy cattle because they are better able to account for environmental effects and accommodate wide variation in milk recording and component sampling. I addition, test days can describe genetic differences in lactation curve (persistency), and genetic effects of parity (rate of maturity). The many genetic estimates resulting for each animal intensifies the need for an appropriate overall lactation performance index. Test day results may be combined into a separate trait index for lactation yield as proposed in most countries implementing test day models. This approach provides continuity with results from a lactation model, but does not fully convey the information contained in test day yields. About 10% of the variation within a lactation is due to differences in persistency. One possible contribution to the economic value of persistency is that cows with relatively more milk late in lactation have lower health costs. In some countries an index for persistency has been proposed. Net present value considerations may make the first lactation economically more important that later ones, but higher yield of mature cows may make later lactations more important. To properly consider all traits, an overall lactation performance index composed of an economic weighting of all components should be developed. Such an index should use correct economic values of all components, and the correct (co)variance among them and should distinguish between breeding goals and traits in the information vector.