On a weighted exponential distribution with a logarithmic weight: Theory and applications

Hassan S Bakouch (1) , Christophe Chesneau (2) , Muhammad Nauman Khan (3)

1 Tanta University
2 LMNO - Laboratoire de Mathématiques Nicolas Oresme
3 KUST - Kohat University of Science and Technology

A new weighted exponential distribution is proposed using a logarithmic weight function. Mathematical properties are studied, including moments, residual life function, order statistics and record values. A parametric estimation is performed using the maximum likelihood method, the weighted least-square method and the maximum product of spacing estimators method. Applications are provided using four real data sets. It is shown that our new distribution works better than a number of classical and recent distributions.

Mots clés: Record values; Survival function; Hazard rate function; Statistical distributions; Parametric estimation

Domaines: Statistiques [math.ST]

Hassan S Bakouch, Christophe Chesneau, Muhammad Nauman Khan. On a weighted exponential distribution with a logarithmic weight: Theory and applications. 2017. ⟨hal-01525502v2⟩