Table 2.
The Empirical and Estimated Standard errors for Estimating βB,1 in Cardiovascular Health Study
| m=1 | m=2 | m=5 | |
|---|---|---|---|
| NCC | 0.05501 (0.04974) n*=536.3 |
0.03948 (0.03722) n*=773.9 |
0.0261 (0.02534) n*=1,384.9 |
| CCH | 0.04767 (0.04524) n*=535.8 |
0.03335 (0.03273) n*=773.4 |
0.0221 (0.02207) n*=1,384.3 |
The empirical standard error of log hazard ratios, the estimated standard error (in parentheses), and the sample size (n*) averaged over 5,000 iterations are shown in each cell. The empirical variance due to both infinite and finite sampling was estimated by adding the estimated variance for the full cohort estimates to the empirical variance resulting from the simulation. The standard error from the full cohort (N=4,388) was 0.0111.