IDS 594 (Sclove) / Fall, 2002 CLASSIFICATION Discriminant Analysis for Fisher Iris Data four variables: petal and sepal length and width for each of 50 cases in each of 3 species MINITAB Worksheet size: 100000 cells ————— 08/21/2002 9:34:51 PM ———————————————————— Welcome to Minitab, press F1 for help. Retrieving project from file: C:\Program Files\MTBWIN\Data\iris.MPJ Discriminant Analysis: C5 versus C1, C2, C3, C4 Linear Method for Response: C5 Predictors: C1 C2 C3 C4 Group 1 2 3 Count 50 50 50 Summary of Classification Put into ....True Group.... Group 1 2 3 1 50 0 0 2 0 48 1 3 0 2 49 Total N 50 50 50 N Correct 50 48 49 Proportion 1.000 0.960 0.980 N = 150 N Correct = 147 Proportion Correct = 0.980 Squared Distance Between Groups 1 2 3 1 0.000 89.864 179.385 2 89.864 0.000 17.201 3 179.385 17.201 0.000 Linear Discriminant Function for Group 1 2 3 Constant -85.21 -71.75 -103.27 C1 23.54 15.70 12.45 C2 23.59 7.07 3.69 C3 -16.43 5.21 12.77 C4 -17.40 6.43 21.08 Summary of Misclassified Observations Observation True Pred Group Squared Probability Group Group Distance 71 ** 2 3 1 130.862 0.000 2 8.670 0.253 3 6.507 0.747 84 ** 2 3 1 149.030 0.000 2 8.439 0.143 3 4.864 0.857 134 ** 3 2 1 133.067 0.000 2 5.253 0.729 3 7.236 0.271