Hello,
If I have the ROC curve from SPSS output, How can I determine the cut off?
Look for the point on the ROC curve that is closest to the top-left corner, which represents perfect classification (100% sensitivity and 100% specificity). This point represents the optimal cutoff point.
It's good idea to consider the area under the ROC curve (AUC),
generally:
AUC > 0.9: Excellent; the model can effectively distinguish between the positive and negative classes.
0.9 > AUC > 0.8: Very good
0.8 > AUC > 0.7: Good
0.7 > AUC > 0.6: Fair
0.6 > AUC: Poor
Samah Mohammed Refay
The selection of a cutoff point depends on the balance between sensitivity and specificity, as well as the impact of false positives and false negatives in your particular use case. Examining the ROC curve and associated metrics will help you identify a suitable cutoff for your model. When selecting the cutoff point from the ROC curve, you have several options available, such as:
- Select the point where "sensitivity + specificity - 1" is maximized
- Select the point closest to the top-left corner of the ROC curve using the "Minimum Distance Approach"
- Consider other relevant metrics based on your research context, such as minimizing false positives or false negatives.
I would compute sensitivity and specificity values for various threshold points. Identify the threshold where the sum of sensitivity and specificity is the highest.
Please check; Youden's Index (Sensitivity + Specificity - 1).
- Badges
- Science topic
- Similar topics
- Biological Science
- Biochemistry
- Enzymology
- Enzymes