RSS is used to obtain a more efficient estimator of the population mean, as compared to the one based on a simple random sample (SRS). This technique is useful when judgment ordering of a SRS of small size can be easily done and fairly accurately, but exact measurement of an observation is difficult and expensive. However, due to the complicated likelihood, parametric estimation with RSS is difficult. The notion of steady-state RSS (SSRSS) introduced in this paper is expected to be useful in this scenario.
展开▼