Table 2.
Data splitting strategies. The differences in strategies are manifested in the distribution of age, sex, and diagnosis between cross-validation folds.
| Splitting by Age/Sex | Splitting by Site | ||||||
|---|---|---|---|---|---|---|---|
| Fold | Age mean (SD) | Number of females (%) | Number of subjects (%MDD) | Fold | Age mean (SD) | Number of females (%) | Number of subjects (%MDD) |
| 1 | 39.98 (17.40) | 322 (60) | 536 (42) | 1 | 50.15 (13.69) | 607 (49) | 1229 (25) |
| 2 | 39.63 (17.81) | 324 (60) | 538 (42) | 2 | 55.01 (12.57) | 294 (51) | 579 (23) |
| 3 | 39.85 (17.57) | 325 (60) | 538 (43) | 3 | 48.66 (13.59) | 315 (57) | 548 (61) |
| 4 | 39.66 (17.94) | 322 (60) | 535 (39) | 4 | 22.90 (4.97) | 299 (72) | 418 (28) |
| 5 | 39.99 (17.56) | 323 (60) | 538 (44) | 5 | 36.72 (19.69) | 272 (60) | 451 (51) |
| 6 | 39.75 (17.25) | 317 (60) | 531 (43) | 6 | 22.53 (10.92) | 293 (65) | 450 (68) |
| 7 | 40.15 (17.89) | 327 (60) | 541 (42) | 7 | 35.94 (12.96) | 295 (71) | 418 (59) |
| 8 | 39.81 (17.93) | 322 (60) | 535 (44) | 8 | 38.85 (12.66) | 348 (81) | 431 (45) |
| 9 | 39.86 (17.73) | 320 (60) | 535 (44) | 9 | 24.79 (16.16) | 203 (54) | 377 (42) |
| 10 | 39.74 (17.80) | 325 (60) | 538 (43) | 10 | 34.95 (15.45) | 301 (65) | 464 (55) |