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

Download:

Current browse context:

math.ST
< prev | next >
new | recent | 2006

Change to browse by:

math
stat
stat.CO

References & Citations

Bookmark

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

Mathematics > Statistics Theory

Title: Accelerating the pool-adjacent-violators algorithm for isotonic distributional regression

Authors: Alexander Henzi, Alexandre Moesching, Lutz Duembgen
(Submitted on 9 Jun 2020 (v1), last revised 9 Feb 2022 (this version, v5))
Abstract: In the context of estimating stochastically ordered distribution functions, the pool-adjacent-violators algorithm (PAVA) can be modified such that the computation times are reduced substantially. This is achieved by studying the dependence of antitonic weighted least squares fits on the response vector to be approximated.
Subjects: Statistics Theory (math.ST); Computation (stat.CO)
MSC classes: 62-08, 62G08, 63G30
Journal reference: Methodology and Computing in Applied Probability 24 (2022), 2633-2645
DOI: 10.1007/s11009-022-09937-2
Cite as: arXiv:2006.05527 [math.ST]
  (or arXiv:2006.05527v5 [math.ST] for this version)

Submission history

From: Lutz Duembgen [view email]
[v1] Tue, 9 Jun 2020 21:58:47 GMT (6kb)
[v2] Tue, 13 Apr 2021 12:06:33 GMT (145kb,D)
[v3] Wed, 21 Jul 2021 12:46:12 GMT (154kb,D)
[v4] Wed, 1 Dec 2021 11:10:47 GMT (155kb,D)
[v5] Wed, 9 Feb 2022 14:44:34 GMT (155kb,D)
Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

Link back to: arXiv, form interface, contact.