Applicability of different test day models and effect on time at first evaluation of Italian Simmental bulls

Abstract

In recent years, test day models (TDm) have been proposed for genetic evaluation programs of dairy cattle. Advantages due to the use of TDm are known (Jensen, 2001). TDms differ for the (co)variance structure among observations: the first proposed model was a single trait repeatability model, whereas later studies investigated multitrait (MT) and random regression (RR) models (Swalve, 1999). RR assume that the shape of the lactation curve is influenced by random genetic and permanent environmental effects and differ for the number of parameters of the function used for fitting the shape of the individual lactation curve (Jamrozik et al., 1997). TDm are commonly considered in terms of accuracy of estimated breeding values but effects due to constraints on the observational structure of data, which result from the application of different models, are not considered. Such constraints are expected to affect the effective number of herds and cows providing records and, as a consequence, time at first evaluation of young bulls. Application of some TDm to a population with small average herd size might be troublesome because of the loss of herds involved in the genetic evaluation program, as reported by Dalzotto (2000) for the Italian Brown population.........

Keywords

• test day models, age at first evaluation, Simmental