Bayesian and non Bayesian inference for extended two parameters model with application in financial and production fields

Abstract

In statistical inference, introducing a probability distribution appropriate for modeling complex, skewed and symmetric datasets plays an important role. This article presents a new method, referred to as the exponential transformed approach, aimed at creating fresh probability models. This method entails transforming independent and identically distributed reduced Kies random variables. This article establishes various statistical and distributional properties of this model. Furthermore, the article employs several estimation methods to estimates the unknown parameter for the proposed model. Simulation experiments are conducted to showcase the effectiveness of the proposed estimators. Additionally, two real-world data analyses demonstrate practical applications in financial and production contexts, and it is shown that the recommended distribution has superior performance compared to other existing models.

Keywords

• Bayesian procedures
• Maximum likelihood estimation
• Maximum product spacing reduced Kies distribution
• Probability density function
• Simulation experiments
• Skewed datasets