Logistic Regression

In this chapter, the authors introduce logistic regression models for a binary outcome under independent and paired designs, which lay the foundation for further study to extend logistic regression models. In general, logistic regression model is also called unconditional logistic regression model. The hypothesis testing methods and goodness‐of‐fit tests for conditional logistic regression analysis are largely similar to those of unconditional logistic regression. Overall, the sample size requirement is greater for logistic regression analysis than ordinary linear regression analysis because the dependent variable is categorical. The goal of logistic regression is to create an equation that can be used to estimate the probability of an event of interest for the dependent outcome based on one or more independent variables. The Wald's test and likelihood ratio test can be used for hypothesis testing of the partial regression coefficient in the logistic regression model.