WSEAS Transactions on Systems
Print ISSN: 1109-2777, E-ISSN: 2224-2678
Volume 24, 2025
Simulation of Simultaneous Confidence Intervals for All Differences of Signal-to-Noise Ratios of Log-Normal Distributions
Authors: , ,
Abstract: A low signal relative to background noise (signal-to-noise ratio: SNR) signifies a challenge in distinguishing the signal from the background noise. Herein, estimates for the simultaneous confidence intervals (SCIs) for the differences between the SNRs of several log-normal (LN) distributions are presented. These are based on the fiducial generalized confidence interval (FGCI), large sample (LS), method of variance estimates recovery (MOVER), and Bayesian (BS) approaches. By using a Monte Carlo simulation study with RStudio programming, all SCIs are compared based on their coverage probabilities and average lengths. The results indicate that LS approach provided shorter average lengths compared to the MOVER approach, the former approach is the most effective for estimating the SCIs for the differences among the SNRs of several LN distributions. Furthermore, these methods were also used to compare the equality of three price-earning ratios: the SET50, SET100, and sSET indexes. In conclusion, the LS approach proved to be the best method and is thus recommended for estimating the SCIs for the differences between the SNRs of multiple LN distributions.
Search Articles
Pages: 52-58
DOI: 10.37394/23202.2025.24.6