AI Paper Atlas

Research map / Learning theory and algorithms / Kernels and regression theory

Can We Trust the Bootstrap in High-dimensions? The Case of Linear Models

Noureddine El Karoui, Elizabeth Purdom

JMLR 2018 · Learning theory and algorithms › Kernels and regression theory · approach: High-dimensional linear regression theory

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
increase p/n from small to 0.5residual bootstrap CIs · Least squares, n=50095% CI error rate: 10-15% at p/n = 0.3, approximately 20% at p/n = 0.5Hurt
increase p/n for pairs bootstrappairs bootstrap CIs · n=500, L1/Huber/L295% CI error rate: error rate drops far below 5%, essentially zero at p/n = 0.5Hurt
use L1 loss with pairs bootstrap at p/n=0.1pairs bootstrap CIs · L1 loss, n=50095% CI error rate: dropping to less than 0.01Hurt
bootstrap from standardized predicted errors instead of raw residuals (ablation)residual bootstrap · L1 and Huber, double exponential errors95% CI error rateImproved
use weighted bootstrap with weights from Equation (6) instead of Poisson(1)pairs bootstrap · L2, n=500, Gaussian design95% CI error rate: 0.051, 0.06, 0.061, 0.057 for kappa .1, .2, .3, .5Improved
slightly misspecify alpha (0.95 instead of 0.92) (ablation)weighted pairs bootstrap · L2, kappa=0.5bootstrap variance: roughly 30% larger than it should beHurt
rescale jackknife variance by 1 - p/n (ablation)jackknife · L2, Gaussian design95% CI error rateImproved
apply the 1 - p/n jackknife rescaling to elliptical design (ablation)jackknife · L2, elliptical design95% CI error rateNo clear effect
apply the jackknife correction with L1 loss (ablation)jackknife · L1, n=100095% CI error rateHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy