A dependent variable is regressed on K independent variables, using sets of sample ob- servations. We denote by SSE the error sum of squares and by R the coefficient of deter- mination for this estimated regression. We want to test the null hypothesis that K, of these independent variables, taken together, do not linearly affect the dependent variable, given that the other (K-K) independent variables are also to be used. Suppose that the regres sion is reestimated, with the K, independent variables of interest excluded. Let SSE* de- note the error sum of squares and R the coefficient of determination for this regression. Show that the statistic for testing our null hypothesis, introduced in Section 13.7, can be expressed as (SSE SSEA SSE)/K, K 1) RR-K-1 1-R K