2 Interpreting the Mx Output

For the young male like-sex pairs, the estimates are $V_E=0.14$, $V_A=0.25$, and $V_D=0.29$. We can calculate standardized variance components by hand, as $V_{E}^{*}=V_E/V_P$, $V_{A}^{*}=V_A/V_P$, and $V_{D}^{*}=V_D/V_P$, where $V_P = V_E + V_A + V_D = 0.6804$ (which can be read directly from the variance in the expected covariance matrix). In this example, random environmental effects account for 20.3% of the variance, additive genetic effects for 36.4% of the variance, and dominance genetic effects for 43.3% of the variance of BMI in young adult males. By $\chi^{2}$ test of goodness-of-fit, our model gives only a marginally acceptable fit to the data ( $\chi^{2}_{3}=7.28, p=0.06$).