We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

math.ST

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Mathematics > Statistics Theory

Title: Finite-sample risk bounds for maximum likelihood estimation with arbitrary penalties

Abstract: The MDL two-part coding $ \textit{index of resolvability} $ provides a finite-sample upper bound on the statistical risk of penalized likelihood estimators over countable models. However, the bound does not apply to unpenalized maximum likelihood estimation or procedures with exceedingly small penalties. In this paper, we point out a more general inequality that holds for arbitrary penalties. In addition, this approach makes it possible to derive exact risk bounds of order $1/n$ for iid parametric models, which improves on the order $(\log n)/n$ resolvability bounds. We conclude by discussing implications for adaptive estimation.
Comments: To appear in IEEE Transactions on Information Theory, 2018
Subjects: Statistics Theory (math.ST); Machine Learning (stat.ML)
MSC classes: 62B10 94A20 94A15 62E17
Cite as: arXiv:1712.10087 [math.ST]
  (or arXiv:1712.10087v1 [math.ST] for this version)

Submission history

From: Jason Klusowski M [view email]
[v1] Fri, 29 Dec 2017 01:15:49 GMT (93kb)

Link back to: arXiv, form interface, contact.