Help on question 1

1. A bank would like to increase the number of customers who use payroll direct deposit as part of the rollout of its new e-banking platform. An anonymized random sample was obtained from previous clients. Data contains the average monthly checking account balanced in (100's) and whether the customer signed up for direct deposit. (Adapted from Exercise 2 on page 470 of our textbook). We would like to use the k-NN classification method to classify the new customer (shown below) as a potential customer who will opt for direct deposit or not. The new customer has an average balance of $320. Use k-NN classification, first with k = 4, then with k = 7. Distances to the new customer from each observation in z-scores are shown below. Justify your answer with calculations. Customer Balance Direct z-balance Distance to new customer 4.3 1 0.8594 1.1 2.27 0.7338 0.93 W N 3.56 0.2786 0.36 4 4.5 1.0163 1.3 4.9 1.3302 1.7 3 0 -0.1609 0.2 4.31 0.8672 1.11 2.01 0 -0.9378 1.19 1.5 -1.3381 1.7 10 1.7 -1.1811 1.5 mean = 3.205 Std dev 1.2742078 New customer 3.2