Comparing two cut-offs

Have a look again at the confusion matrix using the full model and a cut-off of 15%, which is stored in the object conf_matrix_15, and another confusion matrix using a cut-off of 20% and the same model, stored in conf_matrix_20. Which of the following statements is true? Recall:

$$\textrm{Classification accuracy} = \frac{(TP + TN)}{(TP + FP + TN + FN)}$$

$$\textrm{Sensitivity} = \frac{TP}{(TP + FN)}$$

$$\textrm{Specificity} = \frac{TN}{(TN + FP)}$$

Moving from a cut-off of 15% to 20%…

Possible answers

Accuracy increases, sensitivity decreases and specificity increases.

Accuracy increases, sensitivity increases and specificity decreases.

Accuracy increases, and sensitivity and specificity decrease.

Accuracy decreases, and sensitivity and specificity increase.

Exercise

Comparing two cut-offs

Instructions

Possible answers