Suppose that you have a bivariate data set and the correlati

Suppose that you have a bivariate data set and the correlation coefficient for the relationship between x and y is strong. Suppose also that you compute a least-squares regression line that seems to fit the linear pattern in this bivariate data. You decide to plot the observed x-values (along the x-axis) against the residuals (along the y-axis). If the pattern in the data is truly linear, you should find that the residuals:

Solution

The answer is :

The residuals are randomly distributed around the line y-y^=0