The jackknife, the bootstrap, and other resampling plans

By Bradley Efron. 1987

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

Generalized random forests (2016, methodology): “To motivate the bootstrap of little bags, we first note that building confidence intervals via half-sampling—whereby we evaluate an estimator on random halves of the training data to estimate its sampling error—is closely related to the bootstrap (Efron, 1982).”
Trend analysis of climate time series: A review of methods (2019, methodology): “Efron (1979, 1982) collected and synthesized earlier works by others and presented ordinary (i.e., pointwise or block length unity) bootstrap resampling.”
Deep Bayesian Bandits Showdown: An Empirical Comparison of Bayesian Deep Networks for Thompson Sampling (2018, methodology): “We found it surprising that the optimized versions of BootstrappedNN decided to use only q = 2 and q = 3 networks respectively (while we set its value to q = 10 in the manually tuned version, and the extra networks did not improve performance significantly).”
Uncertainty-Aware Reinforcement Learning for Collision Avoidance (2017, methodology): “To obtain uncertainty estimates from the neural network, we leverage uncertainty estimation methods for discriminatively trained neural networks based on a combination of bootstrapping [5] and dropout [28, 7].”
Deep Exploration via Bootstrapped DQN (2016, methodology): “We actually found the simple non-parametric bootstrap with random initialization [5] more eﬀective in our experiments, but the main ideas of this paper would apply with any other approach to uncertainty in DNNs.”
Randomized Prior Functions for Deep Reinforcement Learning (2018, methodology): “Our approach builds on an alternative method for uncertainty in deep RL inspired by the statistical bootstrap [15].”
Generalization and Exploration via Randomized Value Functions (2014, methodology): “An adaptation of RLSVI that instead samples approximate posterior samples for non-linear value function representations, for example via the online bootstrap [5], may help to explore the environment more efficiently.”
Understanding Membership Inferences on Well-Generalized Learning Models (2018, methodology): “To address this issue, we use bootstrap sampling [12] to generate the Figure 4: Generate new features for vulnerable record selection”
Deep Exploration via Randomized Value Functions (2017, methodology): “The most common form of statistical bootstrap uses the sample data as an approximation to its generating distribution (Efron, 1982).”
A diffusion model account of the lexical decision task. (2004, background): “The idea was to represent what would happen with a different random sample of subjects than those that actually participated in the experiment (see Efron, 1982).”
Quo vadis eohippus? The systematics and taxonomy of the early Eocene equids (Perissodactyla) (2002, methodology): “Jackknifing is a statistical technique in which data points are removed and the change in statistical parameters is examined to test the robustness of those results (Efron, 1982).”
Randomized numerical linear algebra: Foundations and algorithms (2020, methodology): “See Efron (1982) for an introduction to resampling methods.”
Hidden Markov processes (2002, methodology): “The parameter of the HMP was estimated using smoothed bootstrap Monte Carlo resampling [96].”