Estimating the Gumbel-Barnett copula parameter of dependence

In this paper, we developed an empirical evaluation of four estimation procedures for the dependence parameter of the Gumbel-Barnett copula obtained from a Gumbel type I distribution. We used the maximum likelihood, moments and Bayesian methods and studied the performance of the estimates, assuming...

Descripción completa

Detalles Bibliográficos
Autores: Portilla Yela, Jennyfer, Tovar Cuevas, José Rafael
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2018
País:Colombia
Institución:Universidad Nacional de Colombia
Repositorio:Repositorio UN
Idioma:español
OAI Identifier:oai:repositorio.unal.edu.co:unal/66493
Acceso en línea:https://repositorio.unal.edu.co/handle/unal/66493
http://bdigital.unal.edu.co/67521/
Access Level:acceso abierto
Palabra clave:51 Matemáticas / Mathematics
31 Colecciones de estadística general / Statistics
bayesiana
copula Gumbel Barnett
correlación
dependencia copula
estimación
simulación
Copula
Dependence
Correlation
Estimation
Bayesian
Simulation
Descripción
Sumario:In this paper, we developed an empirical evaluation of four estimation procedures for the dependence parameter of the Gumbel-Barnett copula obtained from a Gumbel type I distribution. We used the maximum likelihood, moments and Bayesian methods and studied the performance of the estimates, assuming three dependence levels and 20 different sample sizes. For each method and scenario, a simulation study was conducted with 1000 runs and the quality of the estimator was evaluated using four different criteria. A Bayesian estimator assuming a Beta(a,b) as prior distribution, showed the best performance regardless the sample size and the dependence structure.