Improving Upon the Marginal Empirical Distribution Functions when the Copula is Known

Research output: Working paperDiscussion paperOther research output

230 Downloads (Pure)

Abstract

At the heart of the copula methodology in statistics is the idea of separating marginal distributions from the dependence structure. However, as shown in this paper, this separation is not to be taken for granted: in the model where the copula is known and the marginal distributions are completely unknown, the empirical distribution functions are semiparametrically efficient if and only if the copula is the independence copula. Incorporating the knowledge of the copula into a nonparametric likelihood yields an estimation procedure which by simulations is shown to outperform the empirical distribution functions, the amount of improvement depending on the copula. Although the known-copula model is arguably artificial, it provides an instructive stepping stone to the more general model of a parametrically specified copula and arbitrary margins.
Original languageEnglish
Place of PublicationTilburg
PublisherEconometrics
Number of pages25
Volume2008-40
Publication statusPublished - 2008

Publication series

NameCentER Discussion Paper
Volume2008-40

Fingerprint

Empirical Distribution Function
Copula
Marginal Distribution
Nonparametric Likelihood
Copula Models
Dependence Structure
Margin
If and only if
Statistics
Unknown
Methodology
Arbitrary
Model

Keywords

  • independence copula
  • nonparametric maximum likelihood estimator
  • score function
  • semiparametric efficiency
  • tangent space

Cite this

Segers, J. J. J., van den Akker, R., & Werker, B. J. M. (2008). Improving Upon the Marginal Empirical Distribution Functions when the Copula is Known. (CentER Discussion Paper; Vol. 2008-40). Tilburg: Econometrics.
@techreport{950a8cda8f8c43a9a5c288ca22b891da,
title = "Improving Upon the Marginal Empirical Distribution Functions when the Copula is Known",
abstract = "At the heart of the copula methodology in statistics is the idea of separating marginal distributions from the dependence structure. However, as shown in this paper, this separation is not to be taken for granted: in the model where the copula is known and the marginal distributions are completely unknown, the empirical distribution functions are semiparametrically efficient if and only if the copula is the independence copula. Incorporating the knowledge of the copula into a nonparametric likelihood yields an estimation procedure which by simulations is shown to outperform the empirical distribution functions, the amount of improvement depending on the copula. Although the known-copula model is arguably artificial, it provides an instructive stepping stone to the more general model of a parametrically specified copula and arbitrary margins.",
keywords = "independence copula, nonparametric maximum likelihood estimator, score function, semiparametric efficiency, tangent space",
author = "J.J.J. Segers and {van den Akker}, R. and B.J.M. Werker",
note = "Pagination: 25",
year = "2008",
language = "English",
volume = "2008-40",
series = "CentER Discussion Paper",
publisher = "Econometrics",
type = "WorkingPaper",
institution = "Econometrics",

}

Segers, JJJ, van den Akker, R & Werker, BJM 2008 'Improving Upon the Marginal Empirical Distribution Functions when the Copula is Known' CentER Discussion Paper, vol. 2008-40, Econometrics, Tilburg.

Improving Upon the Marginal Empirical Distribution Functions when the Copula is Known. / Segers, J.J.J.; van den Akker, R.; Werker, B.J.M.

Tilburg : Econometrics, 2008. (CentER Discussion Paper; Vol. 2008-40).

Research output: Working paperDiscussion paperOther research output

TY - UNPB

T1 - Improving Upon the Marginal Empirical Distribution Functions when the Copula is Known

AU - Segers, J.J.J.

AU - van den Akker, R.

AU - Werker, B.J.M.

N1 - Pagination: 25

PY - 2008

Y1 - 2008

N2 - At the heart of the copula methodology in statistics is the idea of separating marginal distributions from the dependence structure. However, as shown in this paper, this separation is not to be taken for granted: in the model where the copula is known and the marginal distributions are completely unknown, the empirical distribution functions are semiparametrically efficient if and only if the copula is the independence copula. Incorporating the knowledge of the copula into a nonparametric likelihood yields an estimation procedure which by simulations is shown to outperform the empirical distribution functions, the amount of improvement depending on the copula. Although the known-copula model is arguably artificial, it provides an instructive stepping stone to the more general model of a parametrically specified copula and arbitrary margins.

AB - At the heart of the copula methodology in statistics is the idea of separating marginal distributions from the dependence structure. However, as shown in this paper, this separation is not to be taken for granted: in the model where the copula is known and the marginal distributions are completely unknown, the empirical distribution functions are semiparametrically efficient if and only if the copula is the independence copula. Incorporating the knowledge of the copula into a nonparametric likelihood yields an estimation procedure which by simulations is shown to outperform the empirical distribution functions, the amount of improvement depending on the copula. Although the known-copula model is arguably artificial, it provides an instructive stepping stone to the more general model of a parametrically specified copula and arbitrary margins.

KW - independence copula

KW - nonparametric maximum likelihood estimator

KW - score function

KW - semiparametric efficiency

KW - tangent space

M3 - Discussion paper

VL - 2008-40

T3 - CentER Discussion Paper

BT - Improving Upon the Marginal Empirical Distribution Functions when the Copula is Known

PB - Econometrics

CY - Tilburg

ER -