Regression Analysis for Outcome-Dependent Sampling Design under the Covariate-Adjusted Additive Hazards Model
Table 3
Simulation results when .
Cutpoints
Method
Bias
SD
SE
CP
Bias
SD
SE
CP
(0.3, 0.7)
Oracle
0.0015
0.1096
0.1146
0.955
0.0032
0.2418
0.2365
0.939
Naive
0.0017
0.1100
0.1145
0.955
0.1220
0.1569
0.1509
0.865
Proposed
0.0014
0.1096
0.1146
0.956
0.0080
0.2377
0.2319
0.934
(0.2, 0.8)
Oracle
0.0000
0.0927
0.0983
0.959
0.2050
0.2024
0.940
Naive
0.0002
0.0927
0.0984
0.961
0.1188
0.1275
0.1288
0.865
Proposed
0.0001
0.0928
0.0983
0.959
0.0046
0.2001
0.1982
0.942
(0.3, 0.7)
Oracle
0.0044
0.1269
0.1265
0.950
0.2622
0.2604
0.942
Naive
0.0046
0.1270
0.1268
0.949
0.1162
0.1691
0.1648
0.905
Proposed
0.0042
0.1268
0.1265
0.949
0.0026
0.2556
0.2548
0.948
(0.2, 0.8)
Oracle
0.0051
0.1007
0.1064
0.957
0.2261
0.2170
0.944
Naive
0.0046
0.1015
0.1065
0.955
0.1172
0.1394
0.1386
0.860
Proposed
0.0050
0.1009
0.1064
0.957
0.2226
0.2126
0.943
“Oracle” denotes the oracle estimator calculated based on the true covariate. “Naive” denotes the naive estimator calculated based on the unadjusted covariate. “Proposed” denotes the proposed estimator calculated based on the adjusted covariate.