Regression Analysis for Outcome-Dependent Sampling Design under the Covariate-Adjusted Additive Hazards Model
Table 4
Simulation results when .
Cutpoints
Method
Bias
SD
SE
CP
Bias
SD
SE
CP
(0.3, 0.7)
Oracle
0.0012
0.0810
0.0894
0.971
0.1812
0.1762
0.955
Naive
0.0009
0.0804
0.0895
0.973
0.1154
0.1151
0.1112
0.843
Proposed
0.0011
0.0809
0.0895
0.971
0.0061
0.1784
0.1726
0.956
(0.2, 0.8)
Oracle
0.0746
0.0773
0.955
0.1533
0.1540
0.949
Naive
0.0747
0.0774
0.957
0.1154
0.0995
0.0973
0.791
Proposed
0.0746
0.0773
0.957
0.0055
0.1523
0.1511
0.950
(0.3, 0.7)
Oracle
0.0879
0.0870
0.943
0.0025
0.1836
0.1799
0.935
Naive
0.0878
0.0872
0.943
0.1166
0.1126
0.1127
0.855
Proposed
0.0877
0.0871
0.942
0.0099
0.1791
0.1762
0.941
(0.2, 0.8)
Oracle
0.0018
0.0738
0.0747
0.954
0.1598
0.1537
0.936
Naive
0.0018
0.0743
0.0748
0.955
0.1180
0.0985
0.0960
0.771
Proposed
0.0019
0.0739
0.0747
0.954
0.0013
0.1568
0.1505
0.936
. “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.