Using the Bayesian Adaptive Lasso Variable Selection Method Comparative Study

Abstract

In this research, we highlight the discovery of the theoretical side of the problem of selecting the variables through a method Assessment of adaptive Lasso regularization and parameter estimation of the multiple linear regression model according to the method of Bayes. D. Lasso's adaptive method is a tool or method of contraction that works at the same time on the choice of language. Explanatory

phrases, which have an effect on a variable The response and parameter estimation regression model has high explanatory power with high prediction accuracy A, where a hierarchical model for prior distributions was proposed, and what is new here is that in this research we have developed the model A subtraction of the prior distributions by suggesting that the data variance  has a prior distribution that follows Inverted square sigmoid How much? Accordingly, the subsequent distributions were derived. In addition, the post-distributions were employed in the work of the Gibbs sampling algorithm to generate samples for dependents. of the model, and one simulation experiment was conducted under different values of the sample sizes, where the method was used. The method of the traditional adaptive lasso and the Bayesian lasso method which were proposed by the Researchers Yi and Mallick in 2014 And compare the results with the proposed method. And the results showed, and based on the MMAD and SD standards, that the proposed method is a method that competes with the other methods. exist in terms of estimating performance, a practical application was made of the data that represent the study of the donor model A study of a group of explanatory variables on the variant of iron deficiency in the blood of a group of patients , and it was

revealed through the employment of The proposed method is a method that provides a method for selecting variables, based on the values of Estimate and compare the results with the methods referred to in the experimental aspect. It is worth mentioning that the results obtained are promising and encouraging, and give better results compared to Existing roads.