Answer:
If we decrease the cutoff value of a logistic regression model, then the probability threshold for predicting the positive class decreases. This means that more instances are predicted as positive, which increases the number of True positives and False positives and decreases the number of True negatives and False negatives.
As a result of this change, the following statements are true:
- False positive rate decreases: This is because the false positive rate is the ratio of false positives to the total number of negatives, and decreasing the cutoff value reduces the number of negatives, which in turn reduces the false positive rate.
- Sensitivity increases: This is because sensitivity is the ratio of true positives to the total number of positives, and decreasing the cutoff value increases the number of true positives, which in turn increases sensitivity.
- Specificity decreases: This is because specificity is the ratio of true negatives to the total number of negatives, and decreasing the cutoff value reduces the number of true negatives, which in turn reduces specificity.
Therefore, the correct answer is: False positive rate decreases, Sensitivity increases, Specificity decreases.