TABLE A8.

Single Confounder Inclusion: simulation results for B(X_j) with n = 500 using parametric estimation;

Variable	1	2	3	4	5
Setting 1
Truth	0.3740	0.2360	0.1840	0.1840	0.0230
Estimate	0.3730	0.2260	0.1860	0.1860	0.0290
Bias	0.0000	0.0100	−0.0020	−0.0020	−0.0060
ESE	0.0810	0.0830	0.0620	0.0590	0.0240
ASE	0.0808	0.0787	0.0625	0.0575	0.0250
Setting 2
Truth	0.6240	0.0030	0.3120	0.0200	0.0410
Estimate	0.5980	0.0210	0.3000	0.0350	0.0460
Bias	0.0250	−0.0180	0.0130	−0.0150	−0.0050
ESE	0.0980	0.0200	0.0910	0.0290	0.0360
ASE	0.0988	0.0242	0.0907	0.0348	0.0366
Setting 3
Truth	0.4510	0.1530	0.0100	0.3490	0.0360
Estimate	0.4370	0.1550	0.0240	0.3440	0.0390
Bias	0.0140	−0.0020	−0.0140	0.0050	−0.0030
ESE	0.0670	0.0540	0.0200	0.0740	0.0270
ASE	0.0703	0.0524	0.0230	0.0718	0.0280
Setting 4
Truth	0.3780	0.2300	0.1870	0.1820	0.0240
Estimate	0.3720	0.2290	0.1850	0.1860	0.0290
Bias	0.0060	0.0010	0.0020	−0.0040	−0.0050
ESE	0.0800	0.0840	0.0640	0.0600	0.0240
ASE	0.0819	0.0797	0.0631	0.0586	0.0254

ESE = empirical standard error (standard deviation of estimates across the 1,000 replications); ASE = average standard error (average of standard error estimates across the 1,000 replications obtained using bootstrapping)