Assessing the Performance of Robust Penalized Regression Methods

Abstract

This thesis is concerned about multiple regression models, when a large number of explanatory variables are consider to explain the relationship between a response Y and a set of predictors X1; ::::;XP . In this thesis, we wish to consider other suitable procedures rather than common least squares. We shall be detecting in further thesis, these alternative model building methods can supply preferable prediction accuracy and model that is easy understandable