Parametric Methods for Regression

Regression analysis is a statistical approach which is used to determine the relationship of a dependent (or output) variable with respect to a few independent variables. In this chapter, we focus on the parametric approaches of regression. We discuss about the mathematical modelModels for regression and how the parameters associated with each variables are identified through the minimization of a cost function. Classical parametric approaches include simple linear regression, also known as the ordinary least square (OLS) regression. Following this, other approaches obtained by modifying OLS such as weighted linear regression, stagewise regression, and least angle regression are discussed. Finally, logistic regressionLogistic regression, a parametric classification algorithm, is also discussed.