🔬  Bootstrap

Bootstrap#

Resampling with replacement to estimate the variability of a statistic.

Important

✨ AI-generated content. This page was written with the assistance of an AI language model and is provided as a learning aid. Despite careful review, it may still contain mistakes, omissions, or out-of-date information. Whether you are new to the topic, a team lead, or a senior practitioner, treat it as a starting point rather than an authoritative reference: read it critically and independently verify anything you act on (code, commands, figures, and factual claims) against official documentation and primary sources before relying on it.

What it is#

The bootstrap is a resampling method that draws new samples with replacement from the observed data — each resample the same size as the original, with some points repeated and others omitted. From many such resamples it estimates a statistic’s sampling distribution.

What it’s for#

By recomputing a statistic (a mean, an AUC) across hundreds or thousands of bootstrap resamples, you get its standard error and confidence intervals without assuming a formula or a distribution. That makes it a flexible, non-parametric way to quantify uncertainty.

Where it appears#

The same idea powers bagging (bootstrap aggregating) and random forests, which train each model on a different bootstrap sample to reduce variance. Its main cost is compute — many refits — and it can struggle with very small samples or extreme statistics.


Theme: Model Evaluation & Uncertainty  ·  All terminology



See also

Source article Adapted (context, re-expressed) in our own words from: Bootstrap (insightful-data-lab.com).

Tags: purpose: reference topic: terminology level: intermediate