Skip to content

Instantly share code, notes, and snippets.

@ctufts
Created April 12, 2017 18:24
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
Star You must be signed in to star a gist
Save ctufts/7b8cc5e4a176a91408ca6bc2f8462545 to your computer and use it in GitHub Desktop.
List of different error metrics

List of metrics First Tier:

  • False Positive - Type I
  • False Negative - Type II
  • True Positive
  • True Negative

Second Tier: Primary issues with 2nd tier - not easy to remember and have various names for the same thing: is there a way to prevent this, i.e. create some type of way to help people understand the concept so well that they don't necessarily need to use the specific term - i.e. they will know what it does/what it tells us (which is more important)

  • Accuracy - (TP + TN)/N
  • Precision - TP/(TP + FP) - AKA Positive Predictive Value
  • Specificity
    • TN/(all negative samples (not predicted))
    • AKA True negative Rate
  • Sensitivity
    • AKA: Recall, True Positive Rate
    • Formula:
  • False negative rate
    • Miss Rate
  • False positive rate
    • Fall Out

Third Tier:

  • ROC/AUC
  • Positive Likelihood Ratio
  • Negative Likelihood Ratio
  • Diagnostic Odds Ratio
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment