MAT 2379 Lecture Notes - Lecture 13: Birth Weight, Fetus, Simple Linear Regression
Document Summary
We want to study the association between two numerical variables. Simple regression is the study of the association between a dependent variable (also called the response) y and an independent variable (also called the predictor) x. We will assume that there is a statistical linear association between the response and the predictor. In other words as we produce a scatter plot of y against x, there should be a linear tendency. We want to find a line to describe the response variable y as a linear function of the explanatory variable x and also define a measure to describe the fit. To be able to find a line of best fit, we need a criterion of fit. That is, we need to know what we mean by best fit. There are many ways to define what is best. Here we present one criterion, which is know as the least-squares criterion.