3. A classification analysis was performed on the apple trees data set from the previous problem. (a) Below are the linear classification functions. Linear Discriminant Function for Rootstock Variable 2 3 5 6 Constant -299.98047 -353.15985 -328.49312 -291.81807 -347.54494 -315.79619 y1 314.63998 317.11988 324.58920 307.26008 316.76668 311.30097 y2 59.41719 -63.98083 -65. 15229 -59.37310 -65.82568 -63.06005 y3 149.61051 168.16143 154.91041 147.65234 168.22072 160.62208 y4 -161.17769 -172.64428 -150.35643 -153.38699 -172.85120 -175.47759 Suppose that we had a tree of unknown rootstock with the following values of variables: y1 = 1.07, y2 = 2.526, y3 = 4.58, y4 = 0.666 What type of rootstock would this tree be classified as?(b) Below this is the classification table. Comment on the results. Could it be improved using a quadratic discrimination function? Why or why not? Number of Observations and Percent Classified into Rootstock From Rootstock 1 2 3 4 5 6 Total 1 5 0 0 0 2 8 62.50 0.00 0.00 12.50 0.00 25.00 100.00 2 3 8 0.00 37.50 25.00 12.50 25.00 0.00 100.00 3 0 8 0.00 0.00 75.00 12.50 12.50 0.00 100.00 4 3 0 4 0 0 8 37.50 0.00 12.50 50.00 0.00 0.00 100.00 5 3 0 3 8 0.00 37.50 12.50 0.00 37.50 12.50 100.00 6 2 0 2 8 25.00 0.00 0.00 0.00 25.00 50.00 100.00 Total 10 6 10 7 8 7 48 20.83 12.50 20.83 14.58 16.67 14.58 100.00 Priors 0.16667 0.16667 0.16667 0.16667 0.16667 0.16667 Error Count Estimates for Rootstock The DISCRIM Procedure Test of Homogeneity of Within Covariance Matrices 1 2 3 5 6 Total Rate 0.3750 0.6250 0.2500 0.5000 0.6250 0.5000 0.4792 Chi-Square DF Pr > ChiSq Priors 0.1667 0.1667 0.1667 0. 1667 0.1667 0.1667 44.018035 50 0.7110