{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,14]],"date-time":"2026-05-14T04:08:01Z","timestamp":1778731681026,"version":"3.51.4"},"reference-count":26,"publisher":"Oxford University Press (OUP)","license":[{"start":{"date-parts":[[2026,5,14]],"date-time":"2026-05-14T00:00:00Z","timestamp":1778716800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/academic.oup.com\/pages\/standard-publication-reuse-rights"}],"funder":[{"DOI":"10.13039\/100000001","name":"NSF","doi-asserted-by":"publisher","award":["DMS-2337882"],"award-info":[{"award-number":["DMS-2337882"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>We introduce a new cross-validation (CV) method based on an equicorrelated Gaussian randomization scheme. Our method is well-suited for problems where sample splitting is infeasible, either because the data violate the assumption of independent and identically distributed samples, or because there are insufficient samples to form representative train\u2013test data pairs. In such problems, our method provides a simple, principled, and computationally efficient approach to estimating prediction error, often outperforming standard CV while requiring only a small number of repetitions. Drawing inspiration from recent splitting techniques like data fission and data thinning, our method constructs train\u2013test data pairs using Gaussian randomization. Our main contribution is the introduction of an antithetic Gaussian randomization scheme, involving a carefully designed correlation structure among the randomization variables. We show theoretically that this antithetic construction can eliminate the bias of CV for a broad class of smooth prediction functions, without inflating variance. Through simulations across a range of data types and loss functions, we demonstrate that our estimator outperforms existing methods for prediction error estimation.<\/jats:p>","DOI":"10.1093\/jrsssb\/qkag073","type":"journal-article","created":{"date-parts":[[2026,4,23]],"date-time":"2026-04-23T11:40:57Z","timestamp":1776944457000},"source":"Crossref","is-referenced-by-count":0,"title":["Cross-validation with antithetic Gaussian randomization"],"prefix":"10.1093","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-1608-4216","authenticated-orcid":false,"given":"Sifan","family":"Liu","sequence":"first","affiliation":[{"name":"Department of Statistical Science, Duke University , Durham, NC ,","place":["USA"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Snigdha","family":"Panigrahi","sequence":"additional","affiliation":[{"name":"Department of Statistics, University of Michigan , Ann Arbor, MI ,","place":["USA"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jake A","family":"Soloff","sequence":"additional","affiliation":[{"name":"Department of Statistics, University of Michigan , Ann Arbor, MI ,","place":["USA"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"286","published-online":{"date-parts":[[2026,5,14]]},"reference":[{"issue":"502","key":"2026051323142988200_qkag073-B1","doi-asserted-by":"publisher","first-page":"741","DOI":"10.1080\/01621459.2013.771582","article-title":"The Poisson compound decision problem revisited","volume":"108","author":"Brown","year":"2013","journal-title":"Journal of the American Statistical Association"},{"issue":"4","key":"2026051323142988200_qkag073-B2","doi-asserted-by":"publisher","first-page":"1534","DOI":"10.1214\/23-AOS2283","article-title":"A cross-validation framework for signal denoising with applications to trend filtering, dyadic CART and beyond","volume":"51","author":"Chaudhuri","year":"2023","journal-title":"Annals of Statistics"},{"issue":"2","key":"2026051323142988200_qkag073-B3","doi-asserted-by":"publisher","first-page":"661","DOI":"10.1214\/009053604000001075","article-title":"Multiprocess parallel antithetic coupling for backward and forward Markov chain Monte Carlo","volume":"33","author":"Craiu","year":"2005","journal-title":"Annals of Statistics"},{"issue":"543","key":"2026051323142988200_qkag073-B4","doi-asserted-by":"publisher","first-page":"1551","DOI":"10.1080\/01621459.2023.2165930","article-title":"A scale-free approach for false discovery rate control in generalized linear models","volume":"118","author":"Dai","year":"2023","journal-title":"Journal of the American Statistical Association"},{"issue":"549","key":"2026051323142988200_qkag073-B5","doi-asserted-by":"publisher","first-page":"511","DOI":"10.1080\/01621459.2024.2353948","article-title":"Generalized data thinning using sufficient statistics","volume":"120","author":"Dharamshi","year":"2025","journal-title":"Journal of the American Statistical Association"},{"issue":"467","key":"2026051323142988200_qkag073-B6","doi-asserted-by":"publisher","first-page":"619","DOI":"10.1198\/016214504000000692","article-title":"The estimation of prediction error: Covariance penalties and cross-validation","volume":"99","author":"Efron","year":"2004","journal-title":"Journal of the American Statistical Association"},{"issue":"1","key":"2026051323142988200_qkag073-B7","doi-asserted-by":"publisher","first-page":"342","DOI":"10.1214\/aos\/1176346597","article-title":"Consistency and asymptotic normality of the maximum likelihood estimator in generalized linear models","volume":"13","author":"Fahrmeir","year":"1985","journal-title":"Annals of Statistics"},{"issue":"1","key":"2026051323142988200_qkag073-B8","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1214\/aos\/1176347963","article-title":"Multivariate adaptive regression splines","volume":"19","author":"Friedman","year":"1991","journal-title":"Annals of Statistics"},{"key":"2026051323142988200_qkag073-B9","doi-asserted-by":"publisher","author":"Fry","year":"2023","DOI":"10.48550\/arXiv.2310.10740"},{"key":"2026051323142988200_qkag073-B10","first-page":"226","article-title":"Measuring sample quality with Stein\u2019s method","volume":"1","author":"Gorham","year":"2015","journal-title":"Proceedings of the 29th International Conference on Neural Information Processing Systems"},{"issue":"2","key":"2026051323142988200_qkag073-B11","doi-asserted-by":"publisher","first-page":"3489","DOI":"10.1214\/25-EJS2416","article-title":"Selective inference using randomized group lasso estimators for general models","volume":"19","author":"Huang","year":"2025","journal-title":"Electronic Journal of Statistics"},{"issue":"549","key":"2026051323142988200_qkag073-B12","doi-asserted-by":"publisher","first-page":"135","DOI":"10.1080\/01621459.2023.2270748","article-title":"Data fission: Splitting a single data point","volume":"120","author":"Leiner","year":"2025","journal-title":"Journal of the American Statistical Association"},{"issue":"12","key":"2026051323142988200_qkag073-B13","first-page":"1","article-title":"Selective inference with distributed data","volume":"26","author":"Liu","year":"2025","journal-title":"Journal of Machine Learning Research: JMLR"},{"issue":"57","key":"2026051323142988200_qkag073-B14","first-page":"1","article-title":"Data thinning for convolution-closed distributions","volume":"25","author":"Neufeld","year":"2024","journal-title":"Journal of Machine Learning Research: JMLR"},{"issue":"2","key":"2026051323142988200_qkag073-B15","doi-asserted-by":"publisher","first-page":"5405","DOI":"10.1214\/24-EJS2322","article-title":"Unbiased risk estimation in the normal means problem via coupled bootstrap techniques","volume":"18","author":"Oliveira","year":"2024","journal-title":"Electronic Journal of Statistics"},{"issue":"1","key":"2026051323142988200_qkag073-B16","doi-asserted-by":"publisher","first-page":"361","DOI":"10.1214\/24-EJS2336","article-title":"Unbiased test error estimation in the Poisson means problem via coupled bootstrap techniques","volume":"19","author":"Oliveira","year":"2025","journal-title":"Electronic Journal of Statistics"},{"issue":"4","key":"2026051323142988200_qkag073-B17","doi-asserted-by":"publisher","first-page":"1109","DOI":"10.1093\/biomet\/asae019","article-title":"Exact selective inference with randomization","volume":"111","author":"Panigrahi","year":"2024","journal-title":"Biometrika"},{"issue":"79","key":"2026051323142988200_qkag073-B18","first-page":"1","article-title":"Approximate post-selective inference for regression with the group lasso","volume":"24","author":"Panigrahi","year":"2023","journal-title":"Journal of Machine Learning Research: JMLR"},{"issue":"544","key":"2026051323142988200_qkag073-B19","doi-asserted-by":"publisher","first-page":"2810","DOI":"10.1080\/01621459.2022.2081575","article-title":"Approximate selective inference via maximum likelihood","volume":"118","author":"Panigrahi","year":"2023","journal-title":"Journal of the American Statistical Association"},{"key":"2026051323142988200_qkag073-B20","first-page":"2825","article-title":"Scikit-learn: Machine learning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"Journal of Machine Learning Research: JMLR"},{"issue":"3","key":"2026051323142988200_qkag073-B21","doi-asserted-by":"publisher","first-page":"597","DOI":"10.1093\/biomet\/asac070","article-title":"Splitting strategies for post-selection inference","volume":"110","author":"Rasines","year":"2023","journal-title":"Biometrika"},{"issue":"3","key":"2026051323142988200_qkag073-B22","doi-asserted-by":"publisher","first-page":"332","DOI":"10.1109\/9.119632","article-title":"Multivariate stochastic approximation using a simultaneous perturbation gradient approximation","volume":"37","author":"Spall","year":"1992","journal-title":"IEEE Transactions on Automatic Control"},{"issue":"6","key":"2026051323142988200_qkag073-B23","doi-asserted-by":"publisher","first-page":"1135","DOI":"10.1214\/aos\/1176345632","article-title":"Estimation of the mean of a multivariate normal distribution","volume":"9","author":"Stein","year":"1981","journal-title":"Annals of Statistics"},{"issue":"2","key":"2026051323142988200_qkag073-B24","doi-asserted-by":"publisher","first-page":"763","DOI":"10.1214\/19-AOS1818","article-title":"Prediction error after model search","volume":"48","author":"Tian","year":"2020","journal-title":"Annals of Statistics"},{"issue":"2","key":"2026051323142988200_qkag073-B25","doi-asserted-by":"publisher","first-page":"679","DOI":"10.1214\/17-AOS1564","article-title":"Selective inference with a randomized response","volume":"46","author":"Tian","year":"2018","journal-title":"Annals of Statistics"},{"issue":"2","key":"2026051323142988200_qkag073-B26","doi-asserted-by":"publisher","first-page":"1198","DOI":"10.1214\/12-AOS1003","article-title":"Degrees of freedom in lasso problems","volume":"40","author":"Tibshirani","year":"2012","journal-title":"Annals of Statistics"}],"container-title":["Journal of the Royal Statistical Society Series B: Statistical Methodology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/jrsssb\/advance-article-pdf\/doi\/10.1093\/jrsssb\/qkag073\/68282372\/qkag073.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/jrsssb\/advance-article-pdf\/doi\/10.1093\/jrsssb\/qkag073\/68282372\/qkag073.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,14]],"date-time":"2026-05-14T03:14:41Z","timestamp":1778728481000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/jrsssb\/advance-article\/doi\/10.1093\/jrsssb\/qkag073\/8677829"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,14]]},"references-count":26,"URL":"https:\/\/doi.org\/10.1093\/jrsssb\/qkag073","relation":{},"ISSN":["1369-7412","1467-9868"],"issn-type":[{"value":"1369-7412","type":"print"},{"value":"1467-9868","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,14]]},"article-number":"qkag073"}}