ABSTRACT
Returning to the ideas behind regression we have seen that, given a set of observations in which each observation or measurement may contain a small amount of error, we can obtain the most likely value of a quantity by distributing the residuals on the basis of minimum variance, which means that the sum of the squares of the residuals is a minimum. This is called the principle of least squares. Although the process of calculating the minimum sum of the squares can be applied to any set of residuals, the interpretation of the results assumes that the errors are normally distributed and of equal weight or that appropriate weights have been applied.
