Regression
Regression analysis is a versatile tool that involves fitting a statistical model to a data set to understand relationships between one or more independent or explanatory variable(s) and the outcome or dependent variable, and make predictions. Types of regression range from simple linear regression, describing a linear relationship between two variables where the outcome is continuous, logistic regression which is used for dichotomous outcomes to multivariable models that allow for simultaneous evaluation of multiple exposures and their interactions while controlling for covariates/potential confounders. In this chapter, we will outline the three most commonly used types of regression analysis in translational research: linear, logistic, and Cox proportional hazards regression, which also assess the time to the occurrence of the outcome.
