Mohammad Esmail Dehghan Monfared

One of the statistical inference methods for unknown parameters is Bayesian statistical inference, in whichapriordensity is considered and bycombiningthepriordensity withtheobtained data, we arrive at the posterior density. All Bayesian inference about the unknown parameters is obtained using this posterior density. To obtain a parametric Bayesian inference with respect to the given loss, posterior mathematical hope or posterior mode is used, but in the first case we need to solve an integral and in the second case we need to maximize a function. In many cases, especially when the likelihood function If it does not have a specific shape, it will be very complicated. In many financial data and financial models, the probability function will be complex and have missing values, which can be referred to the random variability model. Usually, the hope maximization method is used to estimate the parameters, which is very dependent on the starting point. Here, the goal is to introduce a new Monte Carlo method and use it to estimate the parameters. This method is called repeated sampling of important points, which is like sampling of important points, but we repeat at each stage of sampling important points.