A mathematical model represented by a linear equation of the form Y = b1X1 + b2X2 + … + bnXn, or by a system of such linear equations. The relationship between the variables in a linear equation is a straight line when plotted on a graph. In multiple regression, Y represents the estimated value of the dependent variable, each of the Xi represents a predictor variable, and the bi are regression coefficients or weights. If they are expressed as standardized regression coefficients, then they are represented by the Greek letter beta (β) instead of b and are called beta weights. See also bootstrapping.