Efficiencies of Several Ranked Set Sampling Techniques versus Simple Random Sampling

• 발행기관: 한국자료분석학회
• 저자명: Seul A Jang Dong Hee Kim
• 간행물 정보: 『Journal of The Korean Data Analysis Society (JKDAS)』Vol.15 No.4, 1743~1753쪽, 전체 11쪽
• 주제분류: 자연과학 > 통계학
• 파일형태: PDF
• 발행일자: 2013.08.30

영문 초록

This study covers the sampling methods based on ranked set sampling (RSS) to estimate the population mean. The purpose of this study is to compare the relative efficiencies (RE) of several sampling methods which we consider in this paper and to find the sampling method with the best efficiency. Including the RSS, the extreme ranked set sampling (ERSS) proposed by Samawi et al. (1996), median ranked set sampling (MRSS) by Muttlak (1998), L ranked set sampling (LRSS) proposed by Al-Nasser (2007), adjusted ranked ordering set sampling (AROSS) proposed by Kim et al. (2007) and folded ranked set sampling (FRSS) proposed by Bani-Mustafa et al. (2011) are introduced and compared with each other in terms of the RE. First of all, we derive the RE for each sampling method analytically. Assuming that the distributions are symmetric about zero, Var(XMRSS) is the smallest and Var(LRSS) is the second smallest, which means that the MRSS has the highest RE except for the AROSS. As a result, comparing the values of RE’s for underlying distributions when the sample size is 3 to 6, we show that the AROSS has the best efficiency followed by the MRSS, LRSS, RSS, ERSS and FRSS for most distributions.

목차

1. Introduction

2. Sampling Methods

3. Relative Efficiencies

4. Numerical Results

5. Conclusion

References

키워드

Ranked Set Sampling Simple Random Sampling Relative Efficiency RE