ABSTRACT
Regression is a method by which we can determine the existence and strength of the relationship between two or more variables. This can be thought of as drawing lines, ideally straight lines, through data points.
Broadly, there are often two separate goals in regression:
Prediction: fitting a predictive model to an observed dataset, then using that model to make predictions about an outcome from a new set of explanatory variables;
Explanation: fit a model to explain the inter-relationships between a set of variables.
The chapter covers the basic concepts around regression such as terminology, residuals, main assumptions, effect modification, interaction, confounding, additive, multiplicative. These concepts are shared with the other two regression models coming up later in the book – logistic and Cox – so this chapter is an important basis for the later ones.
