# Quantifying the Value of a Binary Classifier

In this example, false positives (predicting a customer will default, when in fact they would not) are assigned a cost of $2500. False negatives (predicting a customer will not default, when in fact they do) are assigned a cost of$5000. This enables selection of a threshold that minimizes the average cost per transaction.

### Training Data

Using the training data, the threshold results in 35 true positives, and, therefore $50-35=15$ false negatives. It also resulted in 27 false positives. The total cost when this metric is applied becomes

$$(15)(\frac{\5000}{false\ negative})+(27)(\frac{\2500}{false\ positive})=\142,500$$

$$\frac{\142,500}{200\ events}=\713\ per\ event$$