Answers are grounded in this paper and cite their sources.
Estimating a finite population mean under stratified double-phase systematic sampling requires methods that can use auxiliary information effectively while accounting for the structure of the sampling design. This study introduces two estimators based on harmonic and geometric combinations of auxiliary-variable sample means collected at the first and second sampling phases. Approximate bias and mean squared error (MSE) expressions are developed, and the unknown coefficients are optimized by minimizing the corresponding MSE functions. The theoretical properties of the proposed estimators are investigated, and their performance is compared with six existing estimators through a Monte Carlo simulation involving three artificial finite populations. These populations represent linear association, nonlinear association with asymmetric disturbances, and periodic patterns in the ordering of population units. Three nested sample-size configurations are considered to examine how estimation accuracy changes as the number of sampled units increases. Empirical MSE and percentage relative efficiency are used as performance measures, while graphical representations provide a visual comparison of the competing estimators across population models and sampling configurations. In addition, three real finite populations are considered to assess their performance using observed data. The proposed harmonic- and geometric-based approaches provide alternative ways to incorporate auxiliary information under stratified double-phase systematic sampling, with their relative advantages assessed through theoretical analysis, numerical comparisons, and graphical evaluation.
No takes yet. Share an insight, caveat, or question.
Daraz et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: