Abstract

This paper presents a study of the consistency of MACE results in the Holstein population. Some biases are highlighted when a full pedigree structure is used in the MACE (SD MACE). They are probably due to discrepancies between genetic trends estimated by national evaluations and they were hidden when only male pedigrees were included (S-MGS MACE). These results do not question the advantages of a SD MACE in comparison with a S-MGS MACE, but some improvements are needed. Therefore a new model is proposed, in which the country effects are replaced by country x birth year effects. Results of a first test showed that this new model was able to correct international EBVs for the biases detected before. The Robust MACE would be an easy solution to obtain MACE results robust to over-or- under estimations of genetic trends, which are one of the main potential sources of bias in the national evaluations.

Published in

Interbull bulletin
2013, volume: 47, number: 47, pages: 82-89

Conference

Interbull 2013, Nantes, France, August 23-25 2013

Authors' information