- Sensitivity and specificity: Quantifying whether tests for diseases are any good
- Predictive value: If you test positive/negative, what are your chances?
- Cutoffs and ROCs: Sensitivity vs. specifity, cutoff selection, and ROC curves
- Prevalence: How rare a disease is, and how it affects PPV and NPV
Positive predictive value (PPV), negative predictive value (NPV)
Receiver operating characteristic (ROC) curves
A rare disease has a prevalence near 0.
The more common a disease, the greater the prevalence.
No comments:
Post a Comment