CHAPTER 15 - Linear and Logistic Regression

Abstract

This chapter discusses linear and logistic regression. Linear regression and logistic regression are two of the more frequently used techniques used in statistics at present. These methods are often used because problems, particularly those concerning humans, usually involve several independent variables. Linear regression is one approach that allows multiple independent variables to be used in the analysis. In the linear regression model, the dependent variable is the observed pulmonary function test value and age, race, and sex are the independent variables. Logistic regression is an approach that allows many possible risk factors to be considered simultaneously. In logistic regression, the dependent variable is disease status (presence or absence) and the potential risk factors are included as the independent variables. The chapter presents the methods for examining the relationship between a response or dependent variable and one or more predictor or independent variables. In linear regression, the relationship between a normally distributed response or dependent variable and one or more continuous predictor or independent variables is examined. In logistic regression, the chapter discusses the relationship between a binary dependent variable and one or more independent variables.