FIG. 4.—

MSE, variance, and bias squared for each estimator applied to a full sample of 30 and a reduced sample of 15 individuals, as functions of parametric gene diversity, considering simulated values based on each of the 783 loci. The simulations incorporated 30 individuals in 15 parent–offspring pairs. (A) ${\widehat{H}}_{\text{full}}$. A quadratic regression of bias squared on H (with the constant and linear terms forced to be 0) is given by (7.187 × 10⁻⁵)H², with R² = 0.959. The Spearman correlation coefficient is −0.8364 for H and MSE and −0.8394 for H and variance. (B) ${\tilde{H}}_{\text{full}}$. The Spearman correlation coefficient is −0.8394 for H and MSE and −0.8394 for H and variance. (C) ${\widehat{H}}_{\text{reduced}}$. The Spearman correlation coefficient is −0.8447 for H and MSE and −0.8447 for H and variance. Each point in the graphs is based on 100,000 simulated data sets, and the same simulations were used for all three estimators.