Bi-log-concave Distribution Functions

Lutz Dümbgen, Petro Kolesnyk, Ralf Wilke

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

48 Downloads (Pure)

Resumé

Nonparametric statistics for distribution functions F or densities f=F′ under qualitative shape constraints constitutes an interesting alternative to classical parametric or entirely nonparametric approaches. We contribute to this area by considering a new shape constraint: F is said to be bi-log-concave, if both logF and log(1−F) are concave. Many commonly considered distributions are compatible with this constraint. For instance, any c.d.f. F with log-concave density f=F′ is bi-log-concave. But in contrast to log-concavity of f, bi-log-concavity of F allows for multimodal densities. We provide various characterisations. It is shown that combining any nonparametric confidence band for F with the new shape constraint leads to substantial improvements, particularly in the tails. To pinpoint this, we show that these confidence bands imply non-trivial confidence bounds for arbitrary moments and the moment generating function of F.
OriginalsprogEngelsk
TidsskriftJournal of Statistical Planning and Inference
Vol/bind184
Sider (fra-til)1-17
ISSN0378-3758
DOI
StatusUdgivet - maj 2017

Bibliografisk note

Published online: 23. November 2016

Emneord

  • Hazard
  • Honest confidence region
  • Moment generating function
  • Moments
  • Reverse hazard
  • Shape constraint

Citer dette

Dümbgen, Lutz ; Kolesnyk, Petro ; Wilke, Ralf. / Bi-log-concave Distribution Functions. I: Journal of Statistical Planning and Inference. 2017 ; Bind 184. s. 1-17.
@article{5e144831cb8b4342a6d498224c385044,
title = "Bi-log-concave Distribution Functions",
abstract = "Nonparametric statistics for distribution functions F or densities f=F′ under qualitative shape constraints constitutes an interesting alternative to classical parametric or entirely nonparametric approaches. We contribute to this area by considering a new shape constraint: F is said to be bi-log-concave, if both logF and log(1−F) are concave. Many commonly considered distributions are compatible with this constraint. For instance, any c.d.f. F with log-concave density f=F′ is bi-log-concave. But in contrast to log-concavity of f, bi-log-concavity of F allows for multimodal densities. We provide various characterisations. It is shown that combining any nonparametric confidence band for F with the new shape constraint leads to substantial improvements, particularly in the tails. To pinpoint this, we show that these confidence bands imply non-trivial confidence bounds for arbitrary moments and the moment generating function of F.",
keywords = "Hazard, Honest confidence region, Moment generating function, Moments, Reverse hazard, Shape constraint, Hazard, Honest confidence region, Moment generating function, Moments, Reverse hazard, Shape constraint",
author = "Lutz D{\"u}mbgen and Petro Kolesnyk and Ralf Wilke",
note = "Published online: 23. November 2016",
year = "2017",
month = "5",
doi = "10.1016/j.jspi.2016.10.005",
language = "English",
volume = "184",
pages = "1--17",
journal = "Journal of Statistical Planning and Inference",
issn = "0378-3758",
publisher = "Elsevier BV North-Holland",

}

Bi-log-concave Distribution Functions. / Dümbgen, Lutz; Kolesnyk, Petro; Wilke, Ralf.

I: Journal of Statistical Planning and Inference, Bind 184, 05.2017, s. 1-17.

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

TY - JOUR

T1 - Bi-log-concave Distribution Functions

AU - Dümbgen, Lutz

AU - Kolesnyk, Petro

AU - Wilke, Ralf

N1 - Published online: 23. November 2016

PY - 2017/5

Y1 - 2017/5

N2 - Nonparametric statistics for distribution functions F or densities f=F′ under qualitative shape constraints constitutes an interesting alternative to classical parametric or entirely nonparametric approaches. We contribute to this area by considering a new shape constraint: F is said to be bi-log-concave, if both logF and log(1−F) are concave. Many commonly considered distributions are compatible with this constraint. For instance, any c.d.f. F with log-concave density f=F′ is bi-log-concave. But in contrast to log-concavity of f, bi-log-concavity of F allows for multimodal densities. We provide various characterisations. It is shown that combining any nonparametric confidence band for F with the new shape constraint leads to substantial improvements, particularly in the tails. To pinpoint this, we show that these confidence bands imply non-trivial confidence bounds for arbitrary moments and the moment generating function of F.

AB - Nonparametric statistics for distribution functions F or densities f=F′ under qualitative shape constraints constitutes an interesting alternative to classical parametric or entirely nonparametric approaches. We contribute to this area by considering a new shape constraint: F is said to be bi-log-concave, if both logF and log(1−F) are concave. Many commonly considered distributions are compatible with this constraint. For instance, any c.d.f. F with log-concave density f=F′ is bi-log-concave. But in contrast to log-concavity of f, bi-log-concavity of F allows for multimodal densities. We provide various characterisations. It is shown that combining any nonparametric confidence band for F with the new shape constraint leads to substantial improvements, particularly in the tails. To pinpoint this, we show that these confidence bands imply non-trivial confidence bounds for arbitrary moments and the moment generating function of F.

KW - Hazard

KW - Honest confidence region

KW - Moment generating function

KW - Moments

KW - Reverse hazard

KW - Shape constraint

KW - Hazard

KW - Honest confidence region

KW - Moment generating function

KW - Moments

KW - Reverse hazard

KW - Shape constraint

UR - https://sfx-45cbs.hosted.exlibrisgroup.com/45cbs?url_ver=Z39.88-2004&url_ctx_fmt=info:ofi/fmt:kev:mtx:ctx&ctx_enc=info:ofi/enc:UTF-8&ctx_ver=Z39.88-2004&rfr_id=info:sid/sfxit.com:azlist&sfx.ignore_date_threshold=1&rft.object_id=954926238444

U2 - 10.1016/j.jspi.2016.10.005

DO - 10.1016/j.jspi.2016.10.005

M3 - Journal article

VL - 184

SP - 1

EP - 17

JO - Journal of Statistical Planning and Inference

JF - Journal of Statistical Planning and Inference

SN - 0378-3758

ER -