Table 2.

The generating model is the AE simplex (parameter set 3) with the parameters α₁, α₂ = α₃, β₁, and β₂ = β₃, as shown

αk	βk	AE simplex	ACE simplex	ACE simplex	AE simplex C rank 1
		df = 68	df = 61	df = 60	df = 64
.10	.10	4.39 + 34.01	1.55 + 3.89	1.55 + 3.89	1.55 + 3.89
.10	.15	8.97 + 80.72	3.12 + 8.10	3.12 + 8.10	3.12 + 8.12
.15	.10	4.70 + 36.97	1.60 + 4.02	1.60 + 4.02	1.60 + 4.02
.15	.15	9.55 + 87.71	3.19 + 8.35	3.19 + 8.35	3.19 + 8.35
−.10	.10	3.40 + 24.35	1.45 + 3.86	1.45 + 3.86	1.45 + 3.86
.10	−.10	6.15 + 21.61	5.71 + 18.45	5.71 + 18.45	5.84 + 19.30
−.10	−.10	4.52 + 17.14	3.94 + 14.04	3.95 + 14.05	4.13 + 14.55

The χ² goodness of fit indices are associated with the incorrect models: the AE simplex, the ACE simplex, and the AE simplex with rank 1 C (i.e., σ²[ζ_Ct] = 0, t = 2,3,4). In these models the parameters α_k and β_k (k = 1,2,3) were fixed to zero. The total χ², given NMZ = NDZ = 1000, is broken down into the MZ and the DZ contributions, respectively

The 68 df model is the standard AE simplex with occasion-specific variances σ²[a_t] = 0 & σ²[e_t] = 0. This model is nested under the true model (i.e., AE simplex with Ph->E transmission parameters α_k and β_k). These results are also given in Table 1

The 61 df model is the standard ACE simplex with occasions specific variances σ²[a_t], σ²[e_t], and σ²[c_t]

The 60 df model is the standard ACE simplex with occasions specific variances σ²[a_t], σ²[e_t], and σ²[c_t] = 0

The 64 df model is the standard AE simplex with occasions specific variances σ²[a_t], σ²[e_t], and σ²[c_t] = 0, and a rank one C covariance matrix (i.e., σ²[ζ_Ct] = 0, t = 2,3,4)