← All research spotlights

Bradley Efron

Photo: Ryan K. Morris, National Science and Technology Medals Foundation, Public domain, Wikimedia Commons

Research spotlight · Unit 13

Bradley Efron

Stanford University

Research field
Statistical inference, resampling methods, empirical Bayes

Bradley Efron introduced the bootstrap in 1979: instead of deriving the standard error of a statistic from a model, a computer draws many samples with replacement from the data themselves, recomputes the statistic each time, and reads the standard error off the spread of the results. His paper showed that the older jackknife is a linear approximation to this idea and tried the method on the sample median, a classification error rate, a ratio, and regression coefficients. The unit follows the paper in Section 3 and uses the method for the median bill length of the Palmer penguins and for Galton's correlation. Efron received the International Prize in Statistics 2019 for the bootstrap.

In the masterclass

Section 3, after the bootstrap standard error, the percentile interval, and the two worked examples that use them.

The research is presented in Unit 13: The Bootstrap and Cross-Validation, in a research spotlight box that explains the result at the level of the unit and links it to what you have just learned.

The paper

Efron, B. (1979). Bootstrap Methods: Another Look at the Jackknife. The Annals of Statistics, 7(1), 1‑26. https://doi.org/10.1214/aos/1176344552

The scientists named here are not contributors to the Statistics Masterclass, are not affiliated with it, and have not endorsed it. The program presents and explains their published research, with references to the original papers.