Given a data set where two traits (BWT and TARSUS) are measured in a set of individuals across a number of years, we can partition variance for each trit into components arising from addiive effects, year effect, and residual (unexplained) effects. We can also apply the same partitioning to the covariance between traits. fits a mean for each trait)īWT TARSUS ~ Trait Trait.SEX !r Trait.ANIMAL Trait.YEAR Note that "Trait" is the multivariate equivalent of mu (i.e. If we wanted to test the significance of the genetic covariance here (against a null hypothesis that COVA=0) we could compare the log-likelihood of this model to one in which we modified the covariance matix associated with ANIMAL such that COVA is constrained to equal zero (there is one additional parameter and hence one less DF in the unconstrained model).
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |