Table I.

Simulation results for a continuous endpoint based on 2000 Monte Carlo simulations. The average treatment effect is zero. δ² = σ² = 2. For IPW, a logistic regression including all 20 X variables is fitted to the data. For OPT, all 20 X variables are included in separate linear models for the two arms. SE_M is the standard error based on Monte Carlo samples; SE_A is the average of large sample standard error estimates over the Monte Carlo samples. Cov.Prob. is the coverage probability of the 95% confidence interval based on large-sample normal approximation.

		SEM/SEA (×1000)			Cov. Prob.
		UNADJ	IPW	OPT	UNADJ	IPW	OPT
n=200	k=5	290/282	220/189	204/197	0.945	0.914	0.941
	10	287/282	221/189	207/194	0.943	0.900	0.932
	20	287/283	223/189	214//189	0.949	0.904	0.913
n=500	k=5	176/178	129/124	127/126	0.952	0.943	0.951
	10	178/179	127/124	126/125	0.955	0.941	0.946
	20	174/179	129/124	128/124	0.956	0.938	0.939
n=1000	k=5	125/126	89/89	89/89	0.958	0.949	0.953
	10	130/126	90/88	90/89	0.947	0.942	0.944
	20	126/126	90/88	90/88	0.953	0.945	0.946