Table 3.

Simulation Results When the Second-Phase Sample Selection Depends on Both Y and Z

r	Covariate	SMLE					MLE0		PSE
		Bias	SE	SEE	CP	RE	Bias	SE	Bias	SE
0.0	X	0.005	0.074	0.073	0.952	1.307	0.291	0.096	0.006	0.085
	Z	0.000	0.047	0.047	0.947	1.146	−0.499	0.044	0.000	0.051
0.1	X	0.004	0.070	0.070	0.952	1.220	0.267	0.093	0.004	0.078
	Z	0.000	0.049	0.049	0.945	1.123	−0.556	0.041	0.001	0.052
0.2	X	0.003	0.067	0.067	0.952	1.154	0.254	0.090	0.002	0.072
	Z	0.000	0.052	0.051	0.944	1.106	−0.609	0.039	0.000	0.055
0.3	X	0.003	0.066	0.066	0.950	1.118	0.241	0.089	0.002	0.070
	Z	0.000	0.056	0.055	0.945	1.092	−0.658	0.038	0.000	0.059

NOTE: Bias and SE are, respectively, the empirical bias and standard error of the parameter estimator; SEE is the empirical mean of the standard error estimator; CP is the coverage probability of the 95% confidence interval; RE is the empirical variance of PSE over that of SMLE. Each entry is based on 10,000 replicates.