A comparative analysis of stratified double folded ranked set sampling performance across various distributions
Efficient statistical estimation is crucial for accurate population parameter estimation. This study introduces and evaluates Stratified Double Folded Ranked Set Sampling (SDFRSS), a modified sampling technique designed to enhance estimation efficiency across various probability distributions. Using...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Penerbit Universiti Kebangsaan Malaysia
2025
|
| Online Access: | http://journalarticle.ukm.my/26005/ http://journalarticle.ukm.my/26005/1/SML%2018.pdf |
| Summary: | Efficient statistical estimation is crucial for accurate population parameter estimation. This study introduces and evaluates Stratified Double Folded Ranked Set Sampling (SDFRSS), a modified sampling technique designed to enhance estimation efficiency across various probability distributions. Using Monte Carlo simulations, SDFRSS is compared with Stratified Simple Random Sampling (SSRS), Stratified Ranked Set Sampling (SRSS), and Stratified Median Ranked Set Sampling (SMRSS) based on Mean Squared Error (MSE) and Relative Efficiency (RE) under multiple distributions, including Normal, Student’s t, Uniform, Exponential, Geometric, Gamma, Beta, Weibull, Log-Normal, Logistic, and Chi-Square. The results showed that SDFRSS consistently outperforms SSRS, SRSS, and SMRSS, particularly in skewed and heavy-tailed distributions, by achieving lower MSE and higher efficiency. It effectively reduces estimation errors while maintaining robustness across different sample sizes and stratification structures. However, for some symmetric distributions, SDFRSS does not always yield the lowest MSE, emphasizing the need for distribution-specific selection of sampling methods. Despite increased computational complexity, SDFRSS provides significant gains in precision and efficiency, making it a valuable tool for researchers in fields requiring accurate statistical estimation. Future research should explore its application in high-dimensional data and real-world statistical problems to further establish its practical utility. |
|---|