EconBase
← All papers

From Cross-Validation to SURE: Asymptotic Risk of Tuned Regularized Estimators

Karun Adusumilli, Maximilian Kasy, Ashia Wilson

arXiv 20 Mar 2026 · Mathematics — Statistics Theory

arXiv:2603.20388 · PDF · OpenAlex · Extracted main text

Abstract

We derive the asymptotic risk function of regularized empirical risk minimization (ERM) estimators tuned by $n$-fold cross-validation (CV). The out-of-sample prediction loss of such estimators converges in distribution to the squared-error loss (risk function) of shrinkage estimators in the normal means model, tuned by Stein's unbiased risk estimate (SURE). This risk function provides a more fine-grained picture of predictive performance than uniform bounds on worst-case regret, which are common in learning theory: it quantifies how risk varies with the true parameter. As key intermediate steps, we show that (i) $n$-fold CV converges uniformly to SURE, and (ii) while SURE typically has multiple local minima, its global minimum is generically well separated. Well-separation ensures that uniform convergence of CV to SURE translates into convergence of the tuning parameter chosen by CV to that chosen by SURE.

Citation extraction

20
references
41
in-text mentions
20
distinct cited
1
self-citations
6,035
main-text words

appendix boundary found by appendix_command · 31% of the source is main text. Read the extracted text to check this.

Most heavily cited references

The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.

ReferenceIntensityMentionsSectionsMain text
1Wilson, A., Kasy, M., and Mackey, L (2020) Approximate cross-validation: Guarantees for model assessment and selection self0.8746367%
2Stein, C. M (1981) Estimation of the mean of a multivariate normal distribution0.81142100%
3van der Vaart, A. W (2000) Asymptotic statistics0.66910530%
4Efron, B (2004) The estimation of prediction error: covariance penalties and cross-validation0.64422100%
5James, W. and Stein, C (1961) Estimation with quadratic loss0.51121100%
6Donoho, D. L. and Johnstone, I. M (1995) Adapting to unknown smoothness via wavelet shrinkage0.40511100%
7Golub, G. H., Heath, M., and Wahba, G (1979) Generalized cross-validation as a method for choosing a good ridge parameter0.40511100%
8Hoerl, A. E. and Kennard, R. W (1970) Ridge regression: Biased estimation for nonorthogonal problems0.40511100%
9Li, K.-C (1987) Asymptotic optimality for $c_p$, $c_l$, cross-validation and generalized cross-validation: Discrete index set0.40511100%
10Mallows, C. L (1973) Some comments on $c_p$0.40511100%

Showing the top 10 of 20 scored citations.

Cited by, within the corpus

arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.

Citing paperIntensityMentionsSections
1Assumption-Lean Shrinkage and Model Averaging for Spatial Parameters0.40511