On Entropy-Based Test of Exponentiality in Ranked Set Sampling

When the sampling units can be easily ranked than quantified, ranked set sampling (RSS) is a viable alternative to the traditional simple random sampling (SRS). Much effort has been made for modifying basic RSS protocol with the aim of deriving more efficient estimators of the population attributes....

Full description

Saved in:
Bibliographic Details
Published in:Communications in statistics. Simulation and computation 2015-04, Vol.44 (4), p.979-995
Main Author: Mahdizadeh, M.
Format: Article
Language:English
Subjects:
Computer simulation
Entropy
Estimators
Information theory
Population (statistical)
Random sampling
Ranked set sampling
Samples
Sampling
Test of fit
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:When the sampling units can be easily ranked than quantified, ranked set sampling (RSS) is a viable alternative to the traditional simple random sampling (SRS). Much effort has been made for modifying basic RSS protocol with the aim of deriving more efficient estimators of the population attributes. Entropy has been seminal in developing measures of distributional disparities as a tool for statistical inference. This article is concerned with testing exponentiality based on sample entropy under some RSS-based designs. A simulation study shows that the proposed tests possess good power properties against several alternatives as compared with the ordinary test based on SRS.
ISSN:0361-0918
1532-4141
DOI:10.1080/03610918.2013.791367