Log (Odds)

#statistics #interview

The scale in original Odds is not symmetrical
- The odds of something happening can go from 1 to infinity
- But the odds of something not happening can go from 1 to 0 which is a small space
- This asymmetry makes it difficult to compare the odds

Log Odds Formula

$Log-Odds = l o g \frac{# of Something Happening}{# of Something not Happening}$ $Log-Odds = l o g \frac{p r o b a b i l i t y}{1 - p r o b a b i l i y}$

Related Notes

Linear Regression
Prior Probability
CNN
Area Under Precision Recall Curve (AUPRC)
Multivariable Linear Regression