The exact FDR–FNR frontier and a separable gap under one-factor Gaussian dependence

A mix of human-written and AI-generated textHuman understanding: all partsmath.ST — Statistics Theorystat.TH — Statistics Theory

Contributed by Samuil Petkov ↗

Version 1 / Nov 19, 2026 / CC BY 4.0

Abstract

We determine the exact asymptotic trade-off between the false discovery rate and the false non-discovery rate in a two-group Gaussian model generated by one unobserved common factor of fixed strength. For fixed mixture weight, signal strength, equicorrelation, and false discovery target, the optimum over all randomized full-vector procedures converges to the lower convex envelope of the independent Gaussian threshold curve with effective signal μ/1−ρ\mu/\sqrt{1-\rho}. Thus not observing the factor causes no additional asymptotic loss relative to the factor-revealed experiment. The converse reveals the factor and reduces exactly to an independent experiment. For achievability, factor estimation, a pathwise threshold-crossing bound, and random-denominator control yield an observable optimal rule with finite-sample FDR calibration. If the three model parameters are replaced by consistent estimates, a plug-in rule attains the same frontier with asymptotic FDR control. We also analyze the broad coordinate-local class δn,i=fn,i(Xi,An)\delta_{n,i}=f_{n,i}(X_i,A_n), where the functions may be nonidentical and dimension-dependent and the shared external seed is independent of the model. Every oracle supporting line lying strictly below the two trivial endpoint values has a positive asymptotic gap from this separable class; an analytic parameter choice certifies that such a line exists. For the constrained problem, the gap vanishes at every level exactly when the oracle frontier is the chord joining the reject-none and reject-all rules; otherwise it is strict at every level strictly between the two trivial endpoint levels. For this fixed rank-one Gaussian family, the first result exactly evaluates the compound frontier and the second proves existence of a strict gap from the stated coordinate-local class. Broad dependent model classes and the complete constrained separable frontier remain open.

Provenance statement

OpenAI GPT-5.6 Sol, accessed through ChatGPT and Codex, was used to develop and audit the proofs and numerical checks, and for drafting. Anthropic Claude Opus 5.5, used through Claude Code, was used for editing, an independent check of Sections 6–7, and to derive the gap-dichotomy lemma and theorem and the parameter-continuity lemma and plug-in theorem for unknown parameters. The author reviewed and takes responsibility for every statement and proof.

Tools used

OpenAI
ChatGPTVersion GPT-5.6 Sol
OpenAI
CodexVersion GPT-5.6 Sol
Anthropic
Claude CodeVersion Claude Opus 5.5

References

  1. Yoav Benjamini and Yosef Hochberg. Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal of the Royal Statistical Society: Series B (Methodological) , 57(1):289--300, 1995. doi:10.1111/j.2517-6161.1995.tb02031.x (https://doi.org/10.1111/j.2517-6161.1995.tb02031.x).DOI
  2. Sylvain Delattre and Etienne Roquain. On the false discovery proportion convergence under Gaussian equi-correlation. Statistics & Probability Letters , 81(1):111--115, 2011. doi:10.1016/j.spl.2010.09.025 (https://doi.org/10.1016/j.spl.2010.09.025).DOI
  3. Jianqing Fan and Xu Han. Estimation of the false discovery proportion with unknown dependence. Journal of the Royal Statistical Society: Series B (Statistical Methodology) , 79(4):1143--1164, 2017. doi:10.1111/rssb.12204 (https://doi.org/10.1111/rssb.12204).DOI
  4. Jianqing Fan, Xu Han, and Weijie Gu. Estimating false discovery proportion under arbitrary covariance dependence. Journal of the American Statistical Association , 107(499):1019--1035, 2012. doi:10.1080/01621459.2012.720478 (https://doi.org/10.1080/01621459.2012.720478).DOI
  5. Jianqing Fan, Yuan Ke, Qiang Sun, and Wen-Xin Zhou. FarmTest : Factor-adjusted robust multiple testing with approximate false discovery control. Journal of the American Statistical Association , 114(528):1880--1893, 2019. doi:10.1080/01621459.2018.1527700 (https://doi.org/10.1080/01621459.2018.1527700).DOI
  6. Helmut Finner, Thorsten Dickhaus, and Markus Roters. Dependency and false discovery rate: Asymptotics. The Annals of Statistics , 35(4):1432--1455, 2007. arXiv:0710.3171 (https://arxiv.org/abs/0710.3171), doi:10.1214/009053607000000046 (https://doi.org/10.1214/009053607000000046).arXivDOI
  7. Chlo é Friguet, Maela Kloareg, and David Causeur. A factor model approach to multiple testing under dependence. Journal of the American Statistical Association , 104(488):1406--1415, 2009. doi:10.1198/jasa.2009.tm08332 (https://doi.org/10.1198/jasa.2009.tm08332).DOI
  8. Christopher Genovese and Larry Wasserman. Operating characteristics and extensions of the false discovery rate procedure. Journal of the Royal Statistical Society: Series B (Statistical Methodology) , 64(3):499--517, 2002. doi:10.1111/1467-9868.00347 (https://doi.org/10.1111/1467-9868.00347).DOI
  9. Prasenjit Ghosh and Arijit Chakrabarti. Bayesian model pursuit and near-oracle sparse signal discovery under dependence, 2026. arXiv:2606.22490 (https://arxiv.org/abs/2606.22490).arXiv
  10. Prasenjit Ghosh and Arijit Chakrabarti. Covariance-adaptive residualization and stagewise calibration for dependent multiple testing, 2026. Version 2, July 2026. arXiv:2606.07466 (https://arxiv.org/abs/2606.07466).arXiv
Show all 21 referencesShow fewer references
  1. Ruth Heller and Saharon Rosset. Optimal control of false discovery criteria in the two-group model. Journal of the Royal Statistical Society: Series B (Statistical Methodology) , 83(1):133--155, 2021. doi:10.1111/rssb.12403 (https://doi.org/10.1111/rssb.12403).DOI
  2. Erratum: Optimal control of false discovery criteria in the two-group model. Journal of the Royal Statistical Society Series B: Statistical Methodology, 83(4), 882, 2021. doi:10.1111/rssb.12441 (https://doi.org/10.1111/rssb.12441).DOI
  3. Dan M. Kluger and Art B. Owen. A central limit theorem for the Benjamini--Hochberg false discovery proportion under a factor model. Bernoulli , 30(1):743--769, 2024. doi:10.3150/23-BEJ1615 (https://doi.org/10.3150/23-BEJ1615).DOI
  4. Jeffrey T. Leek and John D. Storey. A general framework for multiple testing dependence. Proceedings of the National Academy of Sciences , 105(48):18718--18723, 2008. doi:10.1073/pnas.0808709105 (https://doi.org/10.1073/pnas.0808709105).DOI
  5. Yutong Nie and Yihong Wu. Large-scale multiple testing: Fundamental limits of false discovery rate control and compound oracle. The Annals of Statistics , 54(2):647--666, 2026. doi:10.1214/25-AOS2581 (https://doi.org/10.1214/25-AOS2581).DOI
  6. Etienne Roquain and Fanny Villers. Exact calculations for false discovery proportion with application to least favorable configurations. The Annals of Statistics , 39(1):584--612, 2011. doi:10.1214/10-AOS847 (https://doi.org/10.1214/10-AOS847).DOI
  7. Rahul Roy and Subir Kumar Bhandari. Asymptotic Bayes' optimality under sparsity for exchangeable dependent multivariate normal test statistics. Statistics & Probability Letters , 207:110030, 2024. doi:10.1016/j.spl.2023.110030 (https://doi.org/10.1016/j.spl.2023.110030).DOI
  8. Wenguang Sun and T. Tony Cai. Large-scale multiple testing under dependence. Journal of the Royal Statistical Society: Series B (Statistical Methodology) , 71(2):393--424, 2009. doi:10.1111/j.1467-9868.2008.00694.x (https://doi.org/10.1111/j.1467-9868.2008.00694.x).DOI
  9. Wensong Wu and Edsel A. Pe ñ a. Bayes multiple decision functions. Electronic Journal of Statistics , 7:1272--1300, 2013. arXiv:1110.0043 (https://arxiv.org/abs/1110.0043), doi:10.1214/13-EJS813 (https://doi.org/10.1214/13-EJS813).arXivDOI
  10. Jichun Xie, T. Tony Cai, and Hongzhe Li. Correction to the paper `` Optimal False Discovery Rate Control for Dependent Data ''. Statistics and Its Interface , 9(1):33--35, 2016. doi:10.4310/SII.2016.v9.n1.e1 (https://doi.org/10.4310/SII.2016.v9.n1.e1).DOI
  11. Jichun Xie, T. Tony Cai, John Maris, and Hongzhe Li. Optimal false discovery rate control for dependent data. Statistics and Its Interface , 4(4):417--430, 2011. doi:10.4310/SII.2011.v4.n4.a1 (https://doi.org/10.4310/SII.2011.v4.n4.a1).DOI

Version history

  1. v1 · 27 pagesSubmitted by Samuil PetkovInitial depositCurrentNov 19, 2026