Copula Integration for Genetic Parameter Estimation in Bivariate Linear Mixed Models
Résumé
In the bivariate gene9c animal model, the es9ma9on methods assume that the traits are gaussian, but in prac9ce, the joint
distribu9on of the two traits can be non-Gaussian. This induces a bias in the gene9c parameters es9ma9on when the
reproducers are non-randomly selected. The aim of this project was to integrate copula func9ons, characterising the joint
distribu9on, in the es9ma9on of gene9c parameters, and so, reduce this bias.
Origine | Fichiers produits par l'(les) auteur(s) |
---|