Genetic parameter estimation for multiple-traits by Henderson's simples method

Abstract

An extension of the Iterative Henderson Simple Method (IHSM) for (co)variance components estimation in a multiple-trait sire model is presented. The method requires there be no missing data for the first trait but allows for missing data in the second and third traits, if in sequential culling. The fixed effects are absorbed into random effects in order to obtain approximate sire solution, and the expectation of the quadratic equation composed of the approximate solutions are equated to their expected values for estimating (co)variance components. The method does not require any inversion of matrices, and proved to be useful for manipulating large data sets; however, solutions are of approximate value. With proper manipulation of the quadratic terms, the relationship among sires may be obtained. Using body weights at weaning and at 12 months of age and scrotal circunference at 12 months of age, variance and (co)variance estimates obtained by IHSM and Henderson's Method 3 were compared.