Table 2.

Accuracy and efficiency in the Bayesian RFA.

	Bias	Cond.	Conv.	b			c
				m(θb) − b	sd(θb)	m(se(θb))	m(θc) − c	sd(θc)	m(se(θc))
CONTINUOUS LATENT VIOLATOR
ρ(T,V) = 0.0	No bias	1	0.95	0.010	0.066	0.001	−0.005	0.073	0.001
	Uniform	2	0.59	0.009	0.062	0.001	0.000	0.078	0.001
	Nonuniform	3	0.93	0.007	0.065	0.001	0.035	0.079	0.001
	Both	4	0.65	0.024	0.065	0.001	0.031	0.087	0.001
ρ(T,V) = 0.5	No bias	5	0.97	−0.011	0.081	0.001	−0.004	0.059	0.001
	Uniform	6	0.93	0.005	0.082	0.001	0.000	0.059	0.001
	Nonuniform	7	0.96	−0.014	0.083	0.001	0.021	0.072	0.001
	Both	8	0.92	0.002	0.083	0.001	0.022	0.070	0.001
CONTINUOUS OBSERVED VIOLATOR
ρ(T,V) = 0.0	No bias	9	1.00	0.006	0.054	0.001	−0.002	0.054	0.001
	Uniform	10	0.91	−0.003	0.047	0.001	0.005	0.057	0.001
	Nonuniform	11	0.92	0.000	0.056	0.001	0.008	0.057	0.001
	Both	12	0.92	−0.008	0.059	0.001	0.015	0.051	0.001
ρ(T,V) = 0.5	No bias	13	0.80	−0.008	0.054	0.001	−0.006	0.047	0.001
	Uniform	14	0.55	−0.006	0.068	0.001	0.008	0.053	0.001
	Nonuniform	15	0.75	−0.002	0.068	0.001	0.013	0.055	0.001
	Both	16	0.52	−0.004	0.063	0.001	0.006	0.046	0.001
DICHOTOMIZED OBSERVED VIOLATOR (AFTER MEDIAN SPLIT OF THE CONTINUOUS OBSERVED VIOLATOR)
ρ(T,V) = 0.0	No bias	17	1.00	−0.002	0.041	0.001	−0.004	0.054	0.001
	Uniform	18	0.99	−0.089	0.046	0.001	0.003	0.055	0.001
	Nonuniform	19	1.00	0.004	0.051	0.001	−0.071	0.063	0.001
	Both	20	0.98	0.087	0.063	0.001	−0.062	0.058	0.001
ρ(T,V) = 0.5	No bias	21	0.94	−0.009	0.051	0.001	0.006	0.056	0.001
	Uniform	22	0.71	−0.112	0.056	0.001	−0.002	0.051	0.001
	Nonuniform	23	0.93	−0.004	0.055	0.001	−0.022	0.071	0.001
	Both	24	0.68	−0.131	0.054	0.001	0.021	0.067	0.001

Cond., Condition; Conv., proportion of converged solutions (of 100 replicates); All summary measures of the parameter estimates are calculated over the converged solutions only.