Nonparametric estimation of simplified vine copula models: comparison of methods

Thomas Nagler, Christian Schellhase, and Claudia Czado. "Nonparametric estimation of simplified vine copula models: comparison of methods." Dependence Modeling 5.1 (2017): 99-120. <http://eudml.org/doc/288474>.

@article{ThomasNagler2017,
abstract = {In the last decade, simplified vine copula models have been an active area of research. They build a high dimensional probability density from the product of marginals densities and bivariate copula densities. Besides parametric models, several approaches to nonparametric estimation of vine copulas have been proposed. In this article, we extend these approaches and compare them in an extensive simulation study and a real data application. We identify several factors driving the relative performance of the estimators. The most important one is the strength of dependence. No method was found to be uniformly better than all others. Overall, the kernel estimators performed best, but do worse than penalized B-spline estimators when there is weak dependence and no tail dependence.},
author = {Thomas Nagler, Christian Schellhase, Claudia Czado},
journal = {Dependence Modeling},
keywords = {B-spline; Bernstein; copula; kernel; nonparametric; simulation; vine},
language = {eng},
number = {1},
pages = {99-120},
title = {Nonparametric estimation of simplified vine copula models: comparison of methods},
url = {http://eudml.org/doc/288474},
volume = {5},
year = {2017},
}

TY - JOUR
AU - Thomas Nagler
AU - Christian Schellhase
AU - Claudia Czado
TI - Nonparametric estimation of simplified vine copula models: comparison of methods
JO - Dependence Modeling
PY - 2017
VL - 5
IS - 1
SP - 99
EP - 120
AB - In the last decade, simplified vine copula models have been an active area of research. They build a high dimensional probability density from the product of marginals densities and bivariate copula densities. Besides parametric models, several approaches to nonparametric estimation of vine copulas have been proposed. In this article, we extend these approaches and compare them in an extensive simulation study and a real data application. We identify several factors driving the relative performance of the estimators. The most important one is the strength of dependence. No method was found to be uniformly better than all others. Overall, the kernel estimators performed best, but do worse than penalized B-spline estimators when there is weak dependence and no tail dependence.
LA - eng
KW - B-spline; Bernstein; copula; kernel; nonparametric; simulation; vine
UR - http://eudml.org/doc/288474
ER -