Nobody Puts Bonferroni in a Corner

Mårten Schultzberg

Home/Archive/arXiv

arXiv · 2604.09256

Nobody Puts Bonferroni in a Corner

Mårten Schultzberg

View original source ↗Read source document ↗Export metadata ↓

Abstract

We argue that Bonferroni correction is a better choice for online experimentation than it is commonly given credit for. The case rests on four considerations. First, it is the simplest broadly implementable FWER-controlling method that produces unconditional simultaneous confidence intervals for every metric. Second, in a well-specified decision framework, guardrail and quality metrics use intersection-union logic and cannot inflate the false positive rate, so the Bonferroni denominator is the number of success metrics only, not the total metric count. Third, it is uniquely tractable for pre-experiment sample size calculations. Fourth, we contextualise the power cost empirically. Drawing on a simulation study and an empirical analysis of 1,296 experiments run on Spotify's experimentation platform, Confidence, we show that the power loss relative to more sophisticated FWER methods depends on both how the correction family is specified and how many metrics are truly non-null. When guardrail metrics are incorrectly included in the family, Holm and Hommel are nearly indistinguishable from Bonferroni. When the family is correctly restricted to success metrics only, they gain roughly 4--5 percentage points in ship rate (the fraction of experiments where the treatment is deployed). When few metrics are truly non-null, the gap narrows to near zero regardless of method.

Explore related subjects

stat.ME

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Mårten Schultzberg. 2026-04-10. Nobody Puts Bonferroni in a Corner. https://arxiv.org/abs/2604.09256

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Modeling cyclostationarity in time series using ASCA

Modern data analysis across diverse disciplines increasingly relies on time series. Many of these datasets exhibit cyclostationarity, where patterns approximately repeat in a regular manner, often across multiple time scales, such as daily, weekly or yearly cycles. In this context, statistical inference is essential to distinguish genuine underlying effects from random variability. While tools like Analysis of Variance (ANOVA) provide such inference, they often lack interpretability and struggle with the complexities of multivariate data. To address these limitations, we propose a unified pipeline for the exploratory analysis of cyclostationary times series using ANOVA Simultaneous Component Analysis (ASCA). ASCA is an extension of ANOVA that is able to work in both univariate and multivariate cases. Combining inference with the visualization capabilities of Principal Component Analysis (PCA), ASCA provides powerful options for interpretability. ASCA's capabilities have been well-established in the analysis of experimental data, but they remain largely unexplored for observational data like time series. Our workflow introduces an algorithmic approach to modeling time-dependent data using ASCA, enabling control over multiple cyclostationary time scales while also accounting for the specific challenges of this type of data, such as autocorrelation. Furthermore, we observed that ASCA provides a better separation of variability across factors than ANOVA in unbalanced designs due to its multivariate nature. We demonstrate the efficacy of this methodology through two real-world case studies: water temperature trends in mountain lakes in Sierra Nevada, Spain, and airborne pollen trends over 30 years recorded in the city of Granada, Spain.

stat.ME↗

arXiv2026

Distributionally balanced sampling designs via tactical configurations

Distributionally balanced sampling designs are low-discrepancy probability designs obtained by minimizing the expected discrepancy between the auxiliary-variable distribution of a random sample and the target population distribution. Existing constructions rely on circular population sequences, which restrict the design space by forcing samples to be contiguous blocks of a sequence. We propose a new construction based on tactical configurations that removes this topological constraint. The resulting designs are fixed-size and have equal inclusion probabilities. We develop both a simple initialization valid for arbitrary population and sample sizes and a spatial initialization that yields a lower initial expected discrepancy, together with a simulated annealing algorithm for optimization within this class. In simulations and empirical examples, the proposed method outperforms state-of-the-art alternatives in terms of distributional fit and balance.

stat.ME↗

arXiv2026

Bayesian Quantile Regression for Misclassified Binary Data with an Application to Spousal Violence Reporting

Survey responses on socially undesirable behaviors, such as self-reported spousal violence, are often subject to underreporting due to social stigma, fear of retaliation, and other reporting pressures. When such data are analyzed using standard econometric models that focus on conditional means, such as probit and logit models, the resulting estimates are likely to be biased and can obscure heterogeneity in covariate effects. To address these challenges, we propose a Bayesian binary quantile regression framework that accounts for misclassification and provides quantile-specific effects for the latent true response. The framework incorporates false-negative and false-positive probabilities to capture reporting errors and employs a novel partially collapsed Gibbs sampler for estimation. We also discuss the computation of covariate effects and marginal likelihood for Bayesian model comparison. Simulation studies under various settings (prior effective sample size, misclassification rates, and prior distribution) show that accounting for misclassification improves inference across quantiles relative to models that ignore reporting errors. We apply the framework to women's self-reported spousal violence and find that underreporting of spousal violence exceeds overreporting across quantiles, while model comparisons using marginal likelihood generally favor the quantile model with misclassification and yield different conclusions about the determinants of reported violence.

stat.ME↗