|
Confidence intervals for the coefficient of variation and the differencebetween coefficients of variation of inverse-gamma distributions |
|---|---|
| รหัสดีโอไอ | |
| Creator | 1. Theerapong Kaewprasert 2. Sa-Aat Niwitpong 3. Suparat Niwitpong |
| Title | Confidence intervals for the coefficient of variation and the differencebetween coefficients of variation of inverse-gamma distributions |
| Publisher | Research and Development Office, Prince of Songkla University |
| Publication Year | 2022 |
| Journal Title | Songklanakarin Journal of Science an Technology (SJST) |
| Journal Vol. | 44 |
| Journal No. | 2 |
| Page no. | 421-432 |
| Keyword | inverse gamma distribution, coefficient of variation, Bayesian method, the highest posterior density, coverage probability |
| URL Website | https://rdo.psu.ac.th/sjst/index.php |
| ISSN | 0125-3395 |
| Abstract | The aim of this study is to establish new confidence intervals for the single coefficient of variation of an inversegamma distribution using Bayesian methods based on the Jeffreys, reference, and uniform priors and compare them with theWald method. The Bayesian methods are constructed with either the credible confidence interval or the highest posterior density(HPD) interval. These concepts were extended to find the difference between the coefficients of variation for two independentinverse-gamma populations. The performances of the proposed confidence intervals were evaluated using coverage probabilitiesand expected lengths via Monte Carlo simulations. The results indicate that the Bayesian HPD interval based on the referenceprior can be recommended for constructing confidence intervals for the coefficient of variation of a single inverse-gammadistribution and the Bayesian HPD interval based on the Jeffreys prior can be recommended for constructing confidence intervalsfor the difference between the coefficients of variation of two inverse-gamma distributions. Rainfall data from northern Thailandwere used to illustrate the efficacies of the proposed methods. |
