An extension of Azzalini’s method

• Filippo Domma ^[1] ; Božidar V. Popović ^[2] ; Saralees Nadarajah ^[3]
  1. [1] University of Calabria

     University of Calabria

     Cosenza, Italia

     
  2. [2] University of Montenegro

     University of Montenegro

     Montenegro

     
  3. [3] University of Manchester

     University of Manchester

     Reino Unido

     

  Mostrar afiliaciones +
• Localización: Journal of computational and applied mathematics, ISSN 0377-0427, Vol. 278, Nº 1 (15 April 2015), 2015, págs. 37-47
• Idioma: inglés
• DOI: 10.1016/j.cam.2014.09.016
• Texto completo no disponible (Saber más ...)
• Resumen

  • The aim of this paper is to extend Azzalini’s method. This extension is done in two stages: consider two dependent and non-identically distributed random variables say X1X1 and X2X2; model the dependence between X1X1 and X2X2 by a copula. To illustrate the new method, we assume X1X1 and X2X2 are exponential random variables. This assumption leads to a new distribution called the Generalized Weighted Exponential Distribution (GWED), a generalization of Gupta and Kundu (2009)’s Weighted Exponential Distribution (WED). Some mathematical properties of the GWED are derived, and its parameters estimated by maximum likelihood. The GWED is applied to biochemical data sets showing its good performance compared to the WED.