Dual Frame Approach for the Estimation of Population Mean Using Ranked Set Sampling and Its Variations

Abstract

In survey sampling, obtaining a complete sampling frame that fully covers the target population is often challenging and costly. Multiple frame sampling provides a cost-effective and practical solution by combining two or more incomplete sampling frames to enhance coverage. This study introduces estimators that integrate Ranked Set Sampling (RSS), Extreme Ranked Set Sampling (ERSS), and Median Ranked Set Sampling (MRSS) within a dual-frame setup for estimating the population mean. The study describes the procedure for selecting samples from each frame, using ranking information, and then combining them to create an improved estimator. The methods presented are especially useful in large-scale surveys where obtaining a complete sampling frame is not feasible. An empirical investigation using real data evaluates the performance of above estimators based on dual-frame simple random sampling (DF-SRS). In addition, a comprehensive simulation study under various population scenarios is conducted to support and validate empirical findings.