[Chapter 9) Classification Confusion Matrix The table below shows a list of 20 observations in a sample data set and an estimated probability of Class 1 membership given by XLMiner. Assume that the cutoff value is 0.50. Actual Class Probability of Probability of Actual Class Class 1 Class 1 0 1.00 1 0.58 0 0.99 0 0.56 1 0.95 1 0.53 0 0.94 1 0.52 0 0.93 0 0.46 1 0.88 1 0.41 0 0.82 0 0.36 0 0.75 0 0.33 1 0.72 0 0.11 0 0.64 1 0.04 Answer the following questions. For fill-in-the-blank questions, write your final answer only (without intermediate steps). Question 39 (12 points) Based on the given information, a classification confusion matrix can be constructed as below: Predicted Class 1 0 Actual Class 1 0 n11 no1 n10 noo In the above classification confusion matrix: n11 = n10 = no1 noo A Round the final answer (if it is not an integer) to the nearest integer. Question 40 (3 points) The Class 1 error rate = A . Use 2- decimal accuracy for the final answer, e.g., 0.12, when necessary. Question 40 (3 points) The Class 1 error rate = A . Use 2- decimal accuracy for the final answer, e.g., 0.12, when necessary. Question 41 (3 points) In general, if we increase the cutoff value, Class 1 error rate may increase or decrease, depending on the sample data will decrease will increase will stay constant Previous Page Next Page Page 27 of 27