letVar(Yi) = σg 2 + σe 2 + σs 2 and Cov(Yi, Yj) = 2 ϕij σg 2 + Isib(i,j) σs 2. In most datasets, including our own, the two models cannot be distinguished statistically [26]. In fact, because Δij = 0.25 and Isib(i,j) = 1 when i and j are full siblings, and Δij = Isib(i,j)= 0 for nearly all other pairs of individuals, it is simple to show that the two models lead to identical predictions of variances and covariances when we set σd 2 = 4 σs 2 and adjust σe 2 appropriately. Although our dataset does not allow us to distinguish between models with only genetic dominance (σd 2 > 0, σs 2 = 0), models with only shared environment (σd 2 = 0, σs 2 > 0), and other intermediate models (σd 2 > 0, σs 2 > 0), comparisons of parameter estimates from these models are informative. In the model with genetic dominance, the quantity H2 = (σd 2 + σg 2)/(σd 2 + σg 2 + σe 2) provides a liberal estimate of the overall impact of genes on the phenotype at hand, whereas in the model attributing any excess similarity between siblings
