tailieunhanh - Báo cáo sinh học: "Estimating variances and covariances for multivariate animal models by restricted maximum likelihood"

Tuyển tập các báo cáo nghiên cứu về sinh học được đăng trên tạp chí sinh học Journal of Biology đề tài: Estimating variances and covariances for multivariate animal models by restricted maximum likelihood | Genet Sei Evol 1991 23 67-83 Elsevier INRA 67 Original article Estimating variances and covariances for multivariate animal models by restricted maximum likelihood K Meyer Edinburgh University Institute of Animal Genetics West Mains Road Edinburgh EH9 3JN UK Received 11 May 1988 accepted 17 December 1990 Summary - Restricted maximum likelihood estimates of variance and covariance components can be obtained by direct maximization of the associated likelihood using standard derivative-free optimization procedures. In general this requires a multi-dimensional search and numerous evaluations of the log likelihood function. Use of this approach for analyses under an animal model has been described for the univariate case. This model includes animals additive genetic merit as random effect and accounts for all relationships between animals. In addition other random factors such as common environmental or maternal genetic effects can be fitted. This paper describes the extension to multivariate analyses allowing for missing records. A numerical example is given and simplifications for specific models are discussed. variance component restricted maximum likelihood animal model additional random effect derivative-free approach multivariate analysis Resume Estimation par le maximum de vraisemblance restreint REML des com-posantes de variance et de covariance pour un modèle animal multicaractères. En se fondant sur le principe du maximum de vraisemblance restreint on peut obtenir les estimations des composantes de variance et de covariance par la recherche directe du maximum de la vraisemblance correspondante au moyen de methodes d optimisation n utilisant pas le calcul de dérivées. En general ceci nécessite une approche multidimensionnelle et de nombreux calculs de la fonction de vraisemblance. I utilisation de cette approche a déjà été décrite dans le cadre d un modèle animal avec un seul caractère. Le modèle considère les valeurs individuelles des animaux comme des effets