Respuesta :

mathmate
It's called the least squares method because the parameters of the regression line are adjusted in such a way that the sum of squares of the difference (between the actual and predicted values) is a minimum.