Higher Order Risk Comparisons of Oracle Calibrated Global Local Priors
Abstract
First-order minimaxity offers limited guidance for choosing among global-local shrinkage priors, since different local-tail distributions can yield the same leading risk. Comparisons at a common global scale may also be misleading, as the scale interacts with the shape of the local prior. We study the worst-case squared-error risks of posterior means in sparse Gaussian normal-means estimation, allowing each prior its own risk-minimizing global scale. For local-variance priors with polynomial tails and logarithmic corrections, we characterize the calibrations that retain first- and second-order minimaxity and derive the asymptotic form of the unrestricted oracle scale. Even after this optimization, differences in risk remain. The polynomial tail index, logarithmic tail exponent and a scale-invariant higher-order correction distinguish priors at successively finer orders. In particular, the horseshoe has smaller asymptotic oracle risk than the horseshoe+, despite the latter's additional logarithmic tail factor. We also establish the calibration accuracy needed to preserve these differences and sufficient conditions for their stability under high-dimensional Gaussian regression designs.
Related
- R2-GP: Variance-Calibrated Priors for Additive Gaussian Processes
- Matching Higher-Order Oracle Complexity for Smooth Monotone Variational Inequalities
- Near-Optimal Higher-Order Oracle Complexity for Convex--Concave Minimax Optimization
- Critical-curve regularity for finite-lifespan frog models via local-to-global comparisons
- Global shadow estimates and Hausdorff measure criteria in higher rank
BibTeX
@misc{han2026higher,
title = {{Higher Order Risk Comparisons of Oracle Calibrated Global Local Priors}},
author = {Dongu Han},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.24352},
eprint = {2610.24352},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.24352},
note = {CC0},
}
RIS
TY - UNPB
TI - Higher Order Risk Comparisons of Oracle Calibrated Global Local Priors
AU - Dongu Han
PY - 2026
DA - 2026/10/17
PB - arXiv
DO - 10.48550/arxiv.2610.24352
UR - https://arxiv.org/abs/2610.24352
AB - First-order minimaxity offers limited guidance for choosing among global-local shrinkage priors, since different local-tail distributions can yield the same leading risk. Comparisons at a common global scale may also be misleading, as the scale interacts with the shape of the local prior. We study the worst-case squared-error risks of posterior means in sparse Gaussian normal-means estimation, allowing each prior its own risk-minimizing global scale. For local-variance priors with polynomial tails and logarithmic corrections, we characterize the calibrations that retain first- and second-order minimaxity and derive the asymptotic form of the unrestricted oracle scale. Even after this optimization, differences in risk remain. The polynomial tail index, logarithmic tail exponent and a scale-invariant higher-order correction distinguish priors at successively finer orders. In particular, the horseshoe has smaller asymptotic oracle risk than the horseshoe+, despite the latter's additional logarithmic tail factor. We also establish the calibration accuracy needed to preserve these differences and sufficient conditions for their stability under high-dimensional Gaussian regression designs.
ER -