Table 2.
Model performance for categorical outcomes as indicated by area under the receiver operator characteristic curve values, and continuous outcomes as indicated by root mean square error and R2 values
| AUC (95% CI) | |||
|---|---|---|---|
| Logistic | Elastic net | Random forest | |
| Binge-eating abstinence | |||
| 10 repeated CV | 0.50 (0.48–0.52) | 0.51 (0.48–0.53) | 0.49 (0.47–0.52) |
| Bootstrap | 0.50 (0.49–0.52) | 0.53 (0.52–0.54) | 0.50 (0.49–0.51) |
| Optimism bootstrap | 0.60 (0.59–0.61) | 0.59 (0.58–0.60) | 0.93 (0.92–0.94) |
| Weight loss ⩾5% | |||
| 10 repeated CV | 0.66 (0.64–0.69) | 0.68 (0.66–0.70) | 0.59 (0.56–0.61) |
| Bootstrap | 0.61 (0.59–0.62) | 0.63 (0.62–0.65) | 0.56 (0.55–0.57) |
| Optimism bootstrap | 0.71 (0.70–0.73) | 0.73 (0.72–0.74) | 0.94 (0.93–0.95) |
| RMSE (95% CI) | R2 (95% CI) | |||||
|---|---|---|---|---|---|---|
| Linear | Elastic net | Random forest | Linear | Elastic net | Random forest | |
| Binge-eating reduction % (log) | ||||||
| 10 repeated CV | 0.42 (0.36–0.48) | 0.29 (0.22–0.36) | 0.28 (0.21–0.35) | 0.04 (0.03–0.05) | 0.04 (0.03–0.05) | 0.05 (0.04–0.06) |
| Bootstrap | 0.53 (0.50–0.57) | 0.35 (0.30–0.40) | 0.35 (0.30–0.40) | 0.012 (0.008–0.015) | 0.012 (0.008–0.015) | 0.011 (0.009–0.013) |
| Optimism bootstrap | 0.51 (0.48–0.55) | 0.48 (0.44–0.52) | 0.37 (0.32–0.42) | 0.009 (0.007–0.013) | 0.049 (0.048–0.052) | 0.634 (0.632–0.636) |
| Eating-disorder psychopathology | ||||||
| 10 repeated CV | 0.81 (0.79–0.84) | 0.73 (0.71–0.75) | 0.73 (0.71–0.76) | 0.18 (0.15–0.21) | 0.27 (0.24–0.30) | 0.25 (0.22–0.28) |
| Bootstrap | 0.91 (0.90–0.93) | 0.74 (0.73–0.75) | 0.75 (0.74–0.76) | 0.12 (0.10–0.13) | 0.23 (0.21–0.24) | 0.21 (0.20–0.23) |
| Optimism bootstrap | 0.81 (0.79–0.82) | 0.74 (0.73–0.75) | 0.61 (0.60–0.62) | 0.17 (0.16–0.19) | 0.23 (0.22–0.24) | 0.63 (0.62–0.64) |
| Weight loss % | ||||||
| 10 repeated CV | 7.39 (7.18–7.59) | 7.19 (6.97–7.41) | 7.07 (6.90–7.24) | 0.12 (0.10–0.14) | 0.12 (0.10–0.15) | 0.05 (0.04–0.07) |
| Bootstrap | 8.29 (8.15–8.43) | 7.61 (7.48–7.74) | 7.30 (7.19–7.42) | 0.07 (0.06–0.08) | 0.056 (0.047–0.065) | 0.021 (0.017–0.026) |
| Optimism bootstrap | 7.34 (7.20–7.48) | 7.08 (6.95–7.21) | 5.88 (5.78–6.02) | 0.10 (0.09–0.11) | 0.10 (0.09–0.11) | 0.592 (0.586–0.594) |
AUC. area under the receiver operator characteristic curve; RMSE, root mean square error, 10 repeated CV, repeated 10-fold cross-validation.
Note: Higher AUC values indicate greater predictive accuracy; lower RMSE values and higher R2 values indicate greater predictive accuracy.