chapter  5
- Regression
Pages 70

Correlation is used to measure the strength of linear relationship between two variables, whereas regression is used to model the response of one variable in terms of one or more other variables. The response variable is known as the dependent (endogenous) variable and the other variables used to predict the response are known as independent (exogenous) variables or predictors. This model can be linear, quadratic, Poisson, nonlinear, and so on, which best ›ts all the data points. Regression is useful to predict the future data points for the response variable, given the predictors.