Table 3.

Sample fit of domain-specific models for alternative copula functions with Gaussian marginals.

	Copula
	Gaussian	Frank	Clayton	Gumbel	Joe
Mobility domain
Log-likelihood	−6656.54	−6665.73	−6727.46	−6669.82	−6736.73
χ2(7) for H0: β3 = β5	29.02***	29.49***	23.82***	33.64***	37.14***
Self-care domain
Log-likelihood	−4221.35	-4212.35	−4248.89	§	§
χ2(7) for H0: β3 = β5	8.31	5.98	5.35
Usual activities domain
Log-likelihood	−6772.96	−6796.04	−6866.11	−6785.64	−6829.65
χ2(7) for H0: β3 = β5	10.87	10.22	10.89	11.23	11.53
Pain/discomfort domain
Log-likelihood	−6148.63	−6148.07	−6190.84	−6147.80	−6199.63
χ2(7) for H0: β3 = β5	29.75***	30.26***	32.71***	29.09***	26.82***
Anxiety/depression domain
Log-likelihood	−6243.59	−6238.86	−6300.55	−6244.72	−6302.70
χ2(7) for H0: β3 = β5	12.05*	8.56	5.10	10.66	11.86

Best-fitting models in bold type (all models have 15 parameters).

^*Statistical significance: 10%.

^***Statistical significance: 1%.

^§No convergence.