Bayesian Approach in Estimation of Shape Parameter of an Exponential Inverse Exponential Distribution
Bashiru Omeiza Sule *
Department of Mathematical Sciences, Kogi State University, Anyigba, Kogi State, Nigeria.
Taiwo Mobolaji Adegoke *
Department of Statistics, University of Ilorin, Kwara State, Nigeria.
*Author to whom correspondence should be addressed.
Abstract
Aims: This study aimed to obtain the shape parameter of an Exponential Inverted Exponential distribution using different prior distributions under different loss functions.
Methodology: The Bayes’ theorem was adopted to obtain the posterior distribution of the shape parameter of an Exponential inverted Exponential distribution for both non-information prior (such as Jeffreys prior, Hartigen prior and Uniform prior) and an informative prior (such as Gamma distribution and chi-square distribution). Different loss functions (such as Entropy loss function, Square error loss function, Al-Bayyati’s loss function and Precautionary loss function) were employed to obtain the estimate parameter of the shape parameter with an assumption that the scale parameter is known.
Results: The posterior distribution of the shape parameter of an Exponential Inverted Exponential distribution follows a Gamma distribution for all the prior distribution in the study. Also the Bayes estimate for the simulated datasets and real life dataset were obtained.
Conclusion: The Bayes’ estimates for different prior distribution under different loss functions are close to the true parameter value of the shape parameter. The estimators are then compared in terms of their Mean Square Error (MSE) which is computed using R programming language. We deduce that the MSE reduces as the sample size (n) increases.
Keywords: Mean square error, exponential inverse exponential distribution, extension of Jeffrey’s prior, Hartigen prior; uniform prior, gamma distribution and chi-square distribution, loss function.