Statistical inferences on different types of Bivariate Pareto distributions

Abstract

In this thesis work, we discuss di erent types of Marshal-Olkin form of bivariate Pareto distribu- tion. We use EM algorithm and Bayesian estimation through slice cum Gibbs sampler to estimate the parameters. The thesis covers some innovative solutions for di erent problems related to im- plementation of EM algorithm and Bayesian estimation. We also analyze the interval estimation of the parameters both in frequentist and bayesian set up. Numerical results are shown to verify the performance of the algorithms. A real-life data analysis is also shown in each study for illustrative purposes