Table 3.
Estimated calibration in the validation set (observed:expected (O:E) ratio) across fifteen different simulated scenarios
| Scenario | Method | |||||
|---|---|---|---|---|---|---|
| Reference: untreated | Ignore treatment | Exclude treated | IPW | IPW, exclude | IPWtrunc exclude | |
| 1 | 1.00 (0.09) | 0.76 (0.07) | 1.01 (0.13) | 0.79 (0.09) | 1.00 (0.13) | 1.00 (0.12) |
| 2 | 1.00 (0.09) | 0.79 (0.07) | 1.00 (0.11) | 0.79 (0.07) | 1.00 (0.11) | 1.00 (0.11) |
| 3 | 1.01 (0.09) | 0.69 (0.07) | 1.00 (0.13) | 0.76 (0.09) | 1.00 (0.13) | 1.00 (0.12) |
| 4 | 1.00 (0.09) | 0.72 (0.07) | 1.01 (0.16) | 0.74 (0.30) | 0.98 (0.44) | 1.00 (0.17) |
| 5 | 1.00 (0.09) | 0.80 (0.08) | 1.00 (0.13) | 0.68 (0.07) | 1.00 (0.10) | 1.00 (0.10) |
| 6 | 1.00 (0.09) | 0.87 (0.08) | 1.01 (0.10) | 0.79 (0.08) | 1.00 (0.10) | 1.00 (0.10) |
| 7 | 1.00 (0.09) | 0.96 (0.09) | 1.01 (0.10) | 0.93 (0.10) | 1.00 (0.10) | 1.00 (0.10) |
| 8 | 1.00 (0.09) | 0.63 (0.06) | 1.01 (0.12) | 0.68 (0.08) | 1.00 (0.13) | 1.00 (0.12) |
| 9 | 1.00 (0.09) | 0.91 (0.08) | 1.01 (0.12) | 0.92 (0.09) | 1.00 (0.13) | 1.00 (0.12) |
| 10 | 1.00 (0.09) | 0.49 (0.06) | 1.00 (0.17) | 0.68 (0.11) | 1.00 (0.20) | 1.00 (0.18) |
| 11 | 1.00 (0.09) | 0.66 (0.07) | 1.00 (0.17) | 0.79 (0.11) | 1.00 (0.20) | 1.00 (0.18) |
| 12 | 1.01 (0.09) | 0.88 (0.08) | 1.01 (0.17) | 0.92 (0.12) | 1.00 (0.20) | 1.00 (0.18) |
| 13 | 1.00 (0.09) | 0.75 (0.07) | 0.90 (0.12) | 0.76 (0.08) | 0.87 (0.12) | 0.88 (0.11) |
| 14 | 1.00 (0.09) | 0.74 (0.07) | 0.70 (0.10) | 0.72 (0.07) | 0.67 (0.10) | 0.67 (0.09) |
| 15 | 1.00 (0.09) | 0.76 (0.07) | 0.39 (0.07) | 0.74 (0.07) | 0.38 (0.07) | 0.38 (0.07) |
Abbreviations: Exclude: exclusion of treated individuals from the analysis; IPW inverse (treatment) probability weighting; IPW trunc IPW with weight truncation at 98th percentile