. Author manuscript; available in PMC: 2013 Jun 4.

Published in final edited form as: J Affect Disord. 2011 Apr 8;132(3):406–412. doi: 10.1016/j.jad.2011.03.008

Table 1E.

Bivariate Model Fitting Statistics for PD

Model Fits Statistics
	Model	−2 Log Likelihood	DF	−2 Log Likelihood Difference	DF Difference	P-value	AIC
0	Full	4653.530	15685	n/a	n/a	n/a	n/a
1	r_A=0	4658.097	15686	4.566	1	0.033	2.566
2	r_A=1	4671.186	15686	17.656	1	0.000	15.656
3	r_C=0	4653.530	15686	0.000	1	1.000	−2.000
4	r_C=1	4653.530	15686	0.000	1	1.000	−2.000
5	r_E=0	4656.847	15686	3.317	1	0.069	1.317
6	r_E=1	4662.219	15686	8.689	1	0.003	6.689
7	c₁ = c₂= r_C=0*	4653.530*	15688*	0.00*	3*	1.00*	−6.00*
8	r_A=c₁ = c₂= r_C=0	4658.536	15689	5.006	4	0.287	−2.994
9	r_A=1, c₁ = c₂= r_C=0	4671.186	15689	17.656	4	0.001	9.656
10	c₁ = c₂= r_C= r_E=0	4656.847	15689	3.317	4	0.506	−4.683
11	c₁ = c₂= r_C= 0, r_E=1	4662.219	15689	8.689	4	0.069	0.689

Abbreviations: A = Additive genetics; AIC = Akaike information criterion; C = Shared environment; E = Unique environment plus error; PD = Panic Disorder as defined in DSM-III-R; NA = not applicable; PG= pathological gambling as defined in DSM-III-R; r_{A =} Correlation between A_PG and A_PD; r_C = Correlation between C_{PG and} C_PD; r_E = Correlation between E_PG and E_PD;

= Best Fitting model