TABLE A5.

Single Confounder Removal: simulation results for λ(−X_j) using nonparametric estimation with n = 2000;

Variable	1	2	3	4	5
Setting 1
Truth	−0.149	−0.097	−0.081	−0.084	−0.019
Estimate	−0.106	−0.057	−0.050	−0.046	0.004
Bias	−0.043	−0.040	−0.030	−0.038	−0.023
ESE	0.015	0.014	0.010	0.009	0.006
Setting 2
Truth	−0.145	−0.007	−0.076	−0.007	−0.015
Estimate	−0.109	0.007	−0.053	0.007	−0.001
Bias	−0.036	−0.013	−0.024	−0.015	−0.014
ESE	0.015	0.003	0.010	0.004	0.004
Setting 3
Truth	−0.187	−0.105	−0.015	−0.142	−0.031
Estimate	−0.122	−0.052	0.020	−0.080	0.003
Bias	−0.065	−0.053	−0.035	−0.062	−0.034
ESE	0.016	0.011	0.010	0.016	0.007
Setting 4
Truth	−0.188	−0.134	−0.125	−0.123	−0.062
Estimate	−0.105	−0.057	−0.049	−0.045	0.004
Bias	−0.084	−0.077	−0.076	−0.078	−0.066
ESE	0.015	0.014	0.010	0.009	0.006

ESE = empirical standard error (standard deviation of estimates across the 1,000 replications)