5. Conclusion
This paper considered several interval estimators for estimating the population signal to noise ratio (SNR). We have used both SRS and RSS techniques to construct the proposed confidence intervals. Since a theoretical comparison is not possible, a simulation study has been conducted to compare the performance of the interval estimators. We have generated data from normal distribution to see the performance of the estimators. Coverage probability and average width are considered as a criterion of a good estimator. When we compare the performance under both sampling techniques, we observed that for large sample sizes, there is not too much difference between the two techniques. However, for small sample sizes, the RSS performed better than the corresponding SRS. We also observed that Methods 1, 3 and 6 performed better than the other methods in most situations. We believe that the findings of the paper will be useful for all practitioners interested in estimating SNR with better accuracy.