In relation to the coefficient of determination ( ) the least squares method plays a role. R^2 indicates how much of the variables variation can be explained by the independent variable. By minimizing squared errors this method contributes to the sum of squares (TSS). A lower error sum results in a fit and higher R^2 showing how well the model explains data variability. Therefore using the least squares method is essential, for evaluating a regression models accuracy and its capacity to clarify data variability