Group of statistical models that models some function of the expected value of the dependent variable as a linear combination of the independent variables. In its simplest form, a linear model specifies the linear relationship between a dependent variable Y, and a set of predictor variables, the X’s, giving:
where b0 is the regression coefficient for the intercept and the bi values the regression coefficients for variables 1 through k computed from the data.
For most statistical analyses, estimates of the linear relationships between variables are adequate to describe the observed data, and for making reasonable predictions for new observations.