Stability via resampling: statistical problems beyond the real line
Abstract
Model averaging techniques based on resampling methods (such as bootstrapping or subsampling) have been utilized across many areas of statistics, often with the explicit goal of promoting stability in the resulting output. We provide a finite-sample theoretical result guaranteeing the stability of bagging when applied to algorithms that return outputs that are not necessarily real-valued—for example, an algorithm that estimates a vector of weights or a density function. We empirically assess the stability of bagging on synthetic and real-world data for a range of problem settings, including causal inference, nonparametric regression, and Bayesian model selection.
Fields of Study & Subjects
Statistical Methods and Inference
Advanced Statistical Process Monitoring
Publication details
- Journal / source
- Electronic Journal of Statistics
- Volume
- 20
- Issue
- 2
- Pages
- 4179–4207
- Publication date
- 1 January 2026
- Language
- EN