For a linear regression equation,the sum of the squared residuals can be computed directly by finding the difference between each Y and its predicted Y,then squaring the difference and adding the squared values.An alternative procedure is to calculate ____. A)r²SS_Y B)(1 - r²)SS_Y C)(1 + r²)SS_Y D)SS_Y/SS_X

Question

Quizplus · Accepted Answer

The alternative procedure is to use the formula:
sum of squared residuals = (1 - r^2) * sum of squared Y deviations
where r is the correlation coefficient and sum of squared Y deviations is calculated as Σ(Y - Ŷ)^2. This formula is derived from the fact that r^2 represents the proportion of variability in Y that is explained by the regression equation, and (1 - r^2) represents the proportion of variability that is unexplained. The unexplained variability is captured by the sum of squared Y deviations. Therefore, multiplying (1 - r^2) by the sum of squared Y deviations gives the sum of squared residuals.

For a Linear Regression Equation,the Sum of the Squared Residuals