Abstract
We compare the instance-wise, finite-sample risks of monotone spectral filters for linear regression, a broad class of estimators including principal component regression (PCR), gradient descent (GD), and ridge regression. We show that PCR dominates all monotone spectral filters: compared to any such filter, the risk of optimally tuned PCR is no bigger by a constant factor for all problems. Furthermore, the dominance is strong if the filter is separated from step functions (e.g., GD and ridge): there exist problem instances for which the risk of PCR is smaller by a polynomial factor in sample size dependence. Our comparison results show that PCR is optimal and thus admissible among monotone filters, significantly extending Wu et al. (2026)'s result that GD strongly dominates ridge. From a technical perspective, we establish new upper and lower bounds for general spectral filters, which are instance-wise sharp when specialized to ridge or GD, recovering or improving the best-known bounds.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Kim, J., Fu, H., Bartlett, P., Lee, J. D., & Wu, J. (2026). Principal Component Regression Dominates all Monotone Spectral Filters for Linear Regression. https://omanscience.com/en/articles/principal-component-regression-dominates-all-monotone-spectral-filters-for-linear-regression
MLA 9
Kim, Juno, et al. "Principal Component Regression Dominates all Monotone Spectral Filters for Linear Regression." https://omanscience.com/en/articles/principal-component-regression-dominates-all-monotone-spectral-filters-for-linear-regression.
Chicago (author–date)
Kim, Juno, Hengyu Fu, Peter Bartlett, Jason D. Lee, and Jingfeng Wu. 2026. "Principal Component Regression Dominates all Monotone Spectral Filters for Linear Regression." https://omanscience.com/en/articles/principal-component-regression-dominates-all-monotone-spectral-filters-for-linear-regression.
Harvard
Kim, J., Fu, H., Bartlett, P., Lee, J. D. and Wu, J. (2026) 'Principal Component Regression Dominates all Monotone Spectral Filters for Linear Regression', Available at: https://omanscience.com/en/articles/principal-component-regression-dominates-all-monotone-spectral-filters-for-linear-regression.
Vancouver
Kim J, Fu H, Bartlett P, Lee JD, Wu J. Principal Component Regression Dominates all Monotone Spectral Filters for Linear Regression. https://omanscience.com/en/articles/principal-component-regression-dominates-all-monotone-spectral-filters-for-linear-regression
IEEE
J. Kim, H. Fu, P. Bartlett, J. D. Lee, and J. Wu, "Principal Component Regression Dominates all Monotone Spectral Filters for Linear Regression," https://omanscience.com/en/articles/principal-component-regression-dominates-all-monotone-spectral-filters-for-linear-regression.