Statistics for Business: Decision Making and Analysis, 3rd Edition

Chapter Summary

Categorical explanatory variables allow regression models to distinguish several groups. Dummy variables encode categorical data as numerical variables for use in regression models. When used as explanatory variables, dummy variables indicate group membership. A multiple regression that includes numerical explanatory variables and a dummy variable that identifies group membership is known as an analysis of covariance. An analysis of covariance adjusts for possible confounding variables when comparing the means of two or more groups. Interaction variables allow the multiple regression to fit different slopes to each group. When comparing groups using regression, it is important to compare the variance of the residuals associated ...