Mixed ranked set sampling design

作者：Haq Abdul^*; Brown Jennifer; Moltchanova Elena; Al Omari Amer Ibrahim

来源：Journal of Applied Statistics, 2014, 41(10): 2141-2156.

DOI：10.1080/02664763.2014.909781

• perfect and imperfect rankings
• mean-squared error
• ranked set sampling
• relative efficiency
• unbiased estimator
• simple random sampling

The main focus of agricultural, ecological and environmental studies is to develop well designed, cost-effective and efficient sampling designs. Ranked set sampling (RSS) is one method that leads to accomplish such objectives by incorporating expert knowledge to its advantage. In this paper, we propose an efficient sampling scheme, named mixed RSS (MxRSS), for estimation of the population mean and median. The MxRSS scheme is a suitable mixture of both simple random sampling (SRS) and RSS schemes. The MxRSS scheme provides an unbiased estimator of the population mean, and its variance is always less than the variance of sample mean based on SRS. For both symmetric and asymmetric populations, the mean and median estimators based on SRS, partial RSS (PRSS) and MxRSS schemes are compared. It turns out that the mean and median estimates under MxRSS scheme are more precise than those based on SRS scheme. Moreover, when estimating the mean of symmetric and some asymmetric populations, the mean estimates under MxRSS scheme are found to be more efficient than the mean estimates with PRSS scheme. An application to real data is also provided to illustrate the implementation of the proposed sampling scheme.

• 出版日期2014-10