On kernel-based estimation of conditional Kendall's tau: finite-distance bounds and asymptotic behavior

Alexis Derumigny; Jean-David Fermanian

On kernel-based estimation of conditional Kendall's tau: finite-distance bounds and asymptotic behavior

Alexis Derumigny, Jean-David Fermanian

Research output: Working paper

25 Downloads (Pure)

Abstract

We study nonparametric estimators of conditional Kendall's tau, a measure of concordance between two random variables given some covariates. We prove non-asymptotic bounds with explicit constants, that hold with high probabilities. We provide "direct proofs" of the consistency and the asymptotic law of conditional Kendall's tau. A simulation study evaluates the numerical performance of such nonparametric estimators.

Original language	English
Place of Publication	Ithaca, NY
Publisher	ArXiv.org
Publication status	Published - 15 Oct 2018
Externally published	Yes

Keywords

Conditional dependence measure
kernel smoothing
Conditional Kendall's tau

Access to Document

Derumigny2019onFinal published version, 435 KB

https://arxiv.org/abs/1810.06234

2 Working paper
1 PhD Thesis - Research external, graduation external

Estimation of a regular conditional functional by conditional U-statistics regression
Derumigny, A., 26 Mar 2019, 35 p.
Research output: Working paper

Open Access
File
Some statistical results in high-dimensional dependence modeling
Derumigny, A., 2019, 304 p.
Research output: Thesis › PhD Thesis - Research external, graduation external

Open Access
File
About Kendall's regression
Derumigny, A. & Fermanian, J.-D., 20 Nov 2018, 37 p.
Research output: Working paper

Open Access
File

Cite this

@techreport{8afa0f1758d040c7b3316c5c84929142,

title = "On kernel-based estimation of conditional Kendall's tau: finite-distance bounds and asymptotic behavior",

abstract = "We study nonparametric estimators of conditional Kendall's tau, a measure of concordance between two random variables given some covariates. We prove non-asymptotic bounds with explicit constants, that hold with high probabilities. We provide {"}direct proofs{"} of the consistency and the asymptotic law of conditional Kendall's tau. A simulation study evaluates the numerical performance of such nonparametric estimators. ",

keywords = "Conditional dependence measure, kernel smoothing, Conditional Kendall's tau",

author = "Alexis Derumigny and Jean-David Fermanian",

note = "29 pages, 4 figures. arXiv admin note: text overlap with arXiv:1802.07613",

year = "2018",

month = oct,

day = "15",

language = "English",

publisher = "ArXiv.org",

type = "WorkingPaper",

institution = "ArXiv.org",

}

TY - UNPB

T1 - On kernel-based estimation of conditional Kendall's tau

T2 - finite-distance bounds and asymptotic behavior

AU - Derumigny, Alexis

AU - Fermanian, Jean-David

N1 - 29 pages, 4 figures. arXiv admin note: text overlap with arXiv:1802.07613

PY - 2018/10/15

Y1 - 2018/10/15

N2 - We study nonparametric estimators of conditional Kendall's tau, a measure of concordance between two random variables given some covariates. We prove non-asymptotic bounds with explicit constants, that hold with high probabilities. We provide "direct proofs" of the consistency and the asymptotic law of conditional Kendall's tau. A simulation study evaluates the numerical performance of such nonparametric estimators.

AB - We study nonparametric estimators of conditional Kendall's tau, a measure of concordance between two random variables given some covariates. We prove non-asymptotic bounds with explicit constants, that hold with high probabilities. We provide "direct proofs" of the consistency and the asymptotic law of conditional Kendall's tau. A simulation study evaluates the numerical performance of such nonparametric estimators.

KW - Conditional dependence measure

KW - kernel smoothing

KW - Conditional Kendall's tau

M3 - Working paper

BT - On kernel-based estimation of conditional Kendall's tau

PB - ArXiv.org

CY - Ithaca, NY

ER -

On kernel-based estimation of conditional Kendall's tau: finite-distance bounds and asymptotic behavior

Abstract

Keywords

Access to Document

Fingerprint

Research output

Estimation of a regular conditional functional by conditional U-statistics regression

Some statistical results in high-dimensional dependence modeling

About Kendall's regression

Cite this