|
Abstract:
|
The variance of a quantitative trait within a population can be decomposed into the sum of environmental and genetic components. The genetic component is then further decomposed into the sum of the variance between families and the variance within the families. Traditionally, in the biological literature, quantitative traits are modeled with the assumption that the within-family genetic variance is identical for each family in the population. In the case where the genetic variance is negligible relative to the environmental variance this is appropriate. However, when this is not the case, a model which allows the within-family genetic variance to depend on the family is to be preferred. We give a theoretical justification for the preferred model, and give results of a simulation study in which we generate distributions of within-family variances based on a major gene model.
|
