Table 2.

Simulation study results with n = 2000; Maximum ASB, Standardized Bias, Standardized Var, and Standardized MSE all multiplied by 100, t^m(Y, X) indicates the MSE-based method that uses outcome information, t^m(X) indicates the MSE-based method that does not use outcome information, t^a(0.20) and t^a(0.10) indicate the ASB-based methods that use a 0.20 and 0.10 threshold, respectively, t^p(0.05) and t^p(0.10) indicate the p-value based methods that use a 0.05 and 0.10 threshold, respectively

	n=2000
	Setting (i)	Setting (ii)	Setting (i)	Setting (ii)
Method	# of Trees		Maximum ASB
Unweighted	–	–	70.45	70.45
TWANG	926	924	12.57	12.68
tm(Y, X)	695	892	13.15	12.66
tm(X)	868	864	12.56	12.68
ta(0.20)	317	319	18.89	18.90
ta(0.10)	907	913	12.81	12.82
tp(0.05)	913	916	12.59	12.69
tp(0.10)	926	924	12.57	12.68
	Standardized Bias		Standardized Var
Unweighted	11.02	36.79	0.22	0.22
TWANG	2.03	6.98	0.28	0.23
tm(Y, X)	2.14	6.98	0.28	0.23
tm(X)	2.04	6.99	0.28	0.23
ta(0.20)	3.15	10.45	0.26	0.21
ta(0.10)	2.07	7.05	0.28	0.23
tp(0.05)	2.03	6.98	0.28	0.23
tp(0.10)	2.03	6.98	0.28	0.23
	Standardized MSE		Design effect
Unweighted	1.43	13.75	1	1
TWANG	0.32	0.72	1.26	1.26
tm(Y, X)	0.32	0.72	1.25	1.26
tm(X)	0.32	0.72	1.26	1.26
ta(0.20)	0.36	1.30	1.18	1.18
ta(0.10)	0.32	0.73	1.26	1.26
tp(0.05)	0.32	0.72	1.26	1.26
tp(0.10)	0.32	0.72	1.26	1.26