Robust regression and machine learning applications

This study, prepared under the name of "Robust Regression and Machine Learning Applications", consists of 5 chapters. In the first part; The aim of the thesis and literature research are included. In the second part; Linear regression, least squares method, weighted least squares method, outlier, breakpoint, leverage point required in robust regression analysis and also the necessary machine learning concepts are included in the applications section. In the third part; There is information about the definition of robust regression analysis and some robust methods of this analysis. In the fourth part; In the third part, the applications using the robust methods given in the chapter are included. The coefficients of the models created in the applications were calculated in two different ways with both algebraic operations and the generated MATLAB code. In addition, models were created by using some prediction algorithms belonging to the supervised learning method of machine learning, apart from robust methods. The success of all the models created were measured using some metrics. The fifth part contains the conclusion part of the thesis. In this section, it is shown that the coefficients of the robust models created for applications give the same values after both algebraic operations and running the generated MATLAB code. Afterwards, the succeses of robust and machine learning models created for each application were compared and comment were made.