Multiple Linear Regression

Multiple linear regression is an extension of simple linear regression in which values on an outcome (Y) variable are predicted from two or more predictor (X) variables. There are three principal objectives of multiple linear regression: (1) to obtain specific predicted values on Y corresponding to specific observed values on the X variables; (2) to determine how well a predetermined set of X variables predict values on Y (i.e., to gauge the predictive strength of this set of predictors, taken together); and (3) to select from a group of X variables a subset that are the “best” predictors of Y. This entry reviews the form of the multiple regression model, assumptions of the analysis, and how to go about selecting and validating a model.

Form of ...

  • Loading...
locked icon

Sign in to access this content

Get a 30 day FREE TRIAL

  • Watch videos from a variety of sources bringing classroom topics to life
  • Read modern, diverse business cases
  • Explore hundreds of books and reference titles