2. In a classical experiment carried out from 1918 to 1934, apple trees of different rootstocks were compared (Andrews and Herzberg 1984, pp. 357-360). The variables analyzed were y1 = trunk girth at 4 years (mm x 100) y2 = extension growth at 4 years (m) y3 = trunk girth at 15 years (mm x 100) y4 = weight of tree above ground at 15 years (1b x 1000) (a) Using the SAS output below, carry out tests of significance for the discriminant functions and find the relative importance of each as in di Di=ldi' i = 1 , ..., s where & are the eigenvalues of the matrix E- H (our textbook calls this W-B). Do these two procedures agree to the number of important discriminant functions? The CANDISC Procedure Eigenvalues of Inv(E)*H CanRsq/(1-CanRsq) Test of HO: The canonical correlations in the current row and all that follow are zero Adjusted Approximate Squared Canonical Canonical Standard Canonical Likelihood Approximate Correlation Correlation Error Correlation Eigenvalue |Difference Proportion Cumulative Ratio F Value Num DF Den DF Pr > 1 0.807623 0.770467 0.050724 0.652255 1.8757 .0850 0.6421 0.6421 0.15400767 4.94 20 130.3 <.0001 the standardized coefficients are below. comment on contribution of variables to separation groups. pooled within-class canonical variable can1 can2 can3 can4 y1 y2 y>