Assessment of Scale Variable of Log-Gamma Distribution: A Bayesian Technique
Abstract
The aim of this paper is to estimate scale parameter ? of Log Gamma distribution by using inverse Gamma and inverse Chi-Square priors. The measures of loss function; Squared error loss function (SELF) and Quadratic loss function (QLF) are compared through the estimates of the scale parameter ? for best results. Wolfram Mathematica 11 was used for the analysis. The results showed that estimates of the scale parameter decrease with increase in sample size tending to the actual value of the scale parameter. This indicates an increase in the estimate of shape parameter under loss functions being considered. However, inverse chi square prior outperforms inverse gamma prior. The posterior risks for QLF are least compared to those under SELF. The Quadratic loss function therefore, appeared to be better than Squared error loss function.
Keywords
References
More Articles from INTERNATIONAL JOURNAL OF APPLIED SCIENCES AND MATHEMATICAL THEORY
Author: Samson Yunusa, Imande Terdon Terlumun,, Etuk Emmanuel Dan, and Micah Michael
Author: Ejes, Valentine, Liberty Ebiwareme
Author: Uwakwe, J. I., Anyanwu, E., Inalegwu, N.
Author: Xi Yang, Wenzhuo Zhang
Author: Olamiji Onafowokan1, Olawale Fadugba2, Deborah Okunola1, Alex Mendy1
