Skip to content
EvalSuite
Documentation menu

Confidence intervals

ROC AUC confidence interval (DeLong)

Implementedconfidence.roc_auc_ci

Definition

Confidence interval for the area under the ROC curve from DeLong's structural-component variance, without resampling.

Formula

AUC ± z·√Var_DeLong(AUC), computed on the logit scale

Range: [0, 1]

Inputs and outputs

  • y_true: binary labels
  • y_prob: scores for the positive class
  • level: confidence level

Returns: result object

Assumptions

No assumptions beyond valid, aligned inputs of the documented types.

Limitations

No metric-specific limitations are documented yet. Interpret the value alongside the task, data, and other metrics.

Python API

PythonSince v0.1.0
import evalsuite as es

es.roc_auc_ci(y_true, y_prob)

References

  1. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves. Biometrics. 1988;44(3):837-845.

Implementation status