# Bootstrap resampling

Source: https://onco.cc/terms/bootstrap/  
OnCo record `bootstrap` (Term). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

Bootstrapping refits a model or recomputes a statistic on many resamples of the data drawn with replacement, and the spread of the results is a confidence interval that needs no formula.

## Summary

Bootstrapping estimates the distribution of an estimator by resampling the data, assigning measures of accuracy to sample estimates (Wikipedia). It gives confidence intervals for a C-index or AUC on a small test set, and, applied to one patient, the spread of predictions across bootstrap-refitted models is a per-prediction reliability signal: a prediction whose tenth-to-ninetieth percentile range is wider than an in-domain cutoff should be flagged rather than reported.

## Fields

- Kind: Term
- Last checked: 2026-09-24
- Also known as: bootstrap; bootstrap resampling; bootstrapped confidence interval; bootstrap confidence intervals; resampling with replacement; P10 to P90 spread
- Tags: cansim-terms

## Notes

- Listed in the CanSim terms map 1.0.0 (docs/onco/terms.json, generated 2026-09-24), CC BY 4.0, attribution: CanSim project, an open, public-data-first cancer foundation-model programme; CanSim page path /terms/bootstrap-resampling.

## Sources

- Wikipedia: https://en.wikipedia.org/wiki/Bootstrapping_(statistics)
- Wikipedia: https://en.wikipedia.org/wiki/Bootstrapping_(statistics)

## Connected records

- terms: [Cancer AI vocabulary (CanSim terms map)](https://onco.cc/terms/cancer-ai-vocabulary/), [Cross-validation and stratified k-fold](https://onco.cc/terms/cross-validation/), [Out-of-distribution detection (Mahalanobis guard)](https://onco.cc/terms/ood-detection/), [Permutation test and the Mann-Whitney U test](https://onco.cc/terms/permutation-test/), [Uncertainty quantification and confidence gates](https://onco.cc/terms/uncertainty-quantification/)

---
JSON: https://onco.cc/api/v1/entities/bootstrap.json