الانحدار الكمي الثنائي البيزي المقيد بدوال الجزاء المعكوسة ( مع تطبيق عملي )

Rp-Bayesian Binary Quantile Regression restricted by inverse penalty functions (with practical application) A .pdf

Abstract
Quantile regression is one of the methods that took a wide place in application in the past two decades, because of the desirable features that these methods enjoy for researchers, as it is not affected by outlires values, meaning that it is considered one of the impregnable methods, and it gives more detail to the effect of explanatory variables on the dependent variable.
There are many researchers who dealt with the issue of quantile regression with binary data using classical as well as Bayesian methods. In this thesis, the researcher tries to build a model for analyzing binary data using the quantile regression method, as the Bayesian method has been adopted, and it is one of the methods that has received wide resonance in recent times because of its accuracy, especially in small samples.
In this thesis, a hierarchical Bayesian method is proposed for parameter estimation and for variable selection. Hierarchical Bayesian variable selection in the context of binary quantile regression. Existing approaches to variable selection in the context of binary classification are sensitive to outliers or heterogeneities and other anomalies. The proposed method in this study overcomes these problems in a highly efficient way.
And by using the effective (Gibbs Sampler) algorithm to estimate the parameters of the model, which is superior to the (Metropolis) algorithm used in previous studies. The results of both the simulation study and real data analysis showed that the proposed Bayesian quantile regression method with inverse Lasso is superior in performance in terms of estimating model parameters related to or affecting the response variable compared to the Bayesian quantile regression method with Lasso and Bayesian quantile regression method.
Republic