Multiple Linear Regression - MLR

See: Regression Line.

Like regular linear regression but with a multiple personality disorder. In regular linear regression, we work with just two variables, an independent variable and a dependent variable. We think (or maybe even know) that changes in the dependent variable are always paired with changes in the dependent variable. The process of linear regression finds the equation of the best line to represent how these two variables change together.

In multiple linear regression, we assume there are more than one independent variables that are related in some way to our dependent variable. The result of a multiple linear regression is an equation of the best fit line which is composed of all of the different independent variables and the numerical degree (given as a slope) to which changes in that variable are paired with changes in the dependent variable. We might use multiple linear regression to try to predict returns on an investment when we believe that market performance, interest rates, prices of materials needed to produce the product, etc. all contribute to changes in the returns.