Table 1.
αk | βk | χ2 | ~N |
---|---|---|---|
ACE | (parameter set 1) | ||
.10 | .10 | 2.096 | 11,700 |
.10 | .15 | 5.034 | 4,700 |
.15 | .10 | 1.976 | 12,100 |
.15 | .15 | 4.885 | 4,800 |
−.10 | .10 | 2.487 | 9,600 |
.10 | −.10 | 5.334 | 4,500 |
−.10 | −.10 | 3.670 | 6,500 |
ACE | (parameter set 2) | ||
.10 | .10 | 1.234 | 19,300 |
.10 | .15 | 2.588 | 9,250 |
.15 | .10 | 1.283 | 18,650 |
.15 | .15 | 2.735 | 8,750 |
−.10 | .10 | 0.907 | 26,350 |
.10 | −.10 | 6.649 | 3,600 |
−.10 | −.10 | 3.892 | 6,100 |
AE | (parameter set 3) | ||
.10 | .10 | 38.408 | 620 |
.10 | .15 | 89.690 | 260 |
.15 | .10 | 41.679 | 580 |
.15 | .15 | 97.266 | 240 |
−.10 | .10 | 27.754 | 860 |
.10 | −.10 | 27.756 | 860 |
−.10 | −.10 | 21.653 | 1,100 |
The χ2 equals the non-centrality parameter times N (N = NMZ (1000) + NDZ (1000)) obtained by fitting the model with the Ph->E transmission parameters (α1, α 2 = α3, β1, β2 = β3) fixed to zero. The approximate sample size (~N) required is based on a power calculation given the type I error probability of alpha = 0.05 and df = 4, and an equal number of MZ and DZ twin pairs (N = Nmz + Ndz)