In simultaneous models traits fully interact among themselves at the phenotypic scale. In a recursive model trait 1 affects trait 2 but the opposite does not happen. For the recursive models it is possible to fit an equivalent model in which traits are included as "fixed effects". A recursive model can be applied to the influence of milk yield on fertility in cattle. From simulations under several scenarios, it is shown that ignoring the recursive relationship leads to a strong overestimation of the genetic correlation, whereas assuming it when it is false leads to an underestimation and regression coefficients different from zero. Therefore, it is important to choose a good model. However, maximum likelihoods were almost always the same in both models. This implies that for a recursive model there are quasi-equivalent non recursive models equally likely. The model could be selected by a priori biological reasons.

A. Legarra, C. Robert-Granie

Proceedings of the World Congress on Genetics Applied to Livestock Production, Volume , , 24.07, 2006
Download Full PDF BibTEX Citation Endnote Citation Search the Proceedings

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.