Equations · Machine learning
Logistic regression
The probability of a yes-or-no outcome as an S-shaped function of a predictor.
p = 1 / (1 + e^(−(β₀ + β₁x)))
Aethel says
p equals 1 over the quantity 1 plus e to the power of negative the quantity beta sub zero plus beta sub one x
What it says
A straight line in x, β₀ plus β₁ x, is passed through the sigmoid so the result always lies between 0 and 1. Each unit increase in x multiplies the odds of the outcome by e to the power of β₁.
Symbol by symbol
| Symbol | Say it | Means |
|---|---|---|
| p | p | the probability that the outcome is yes |
| β₀ | beta sub zero, or beta naught | the intercept |
| β₁ | beta sub one | the coefficient of x |
More onβBeta
Where you meet it
Statistics, epidemiology, credit scoring and machine learning.
Questions
How do you say Logistic regression out loud?
p = 1 / (1 + e^(−(β₀ + β₁x))) is read: “p equals 1 over the quantity 1 plus e to the power of negative the quantity beta sub zero plus beta sub one x.” That is exactly what Aethel says when it meets the formula in a book.
What does p mean in Logistic regression?
p (said “p”) is the probability that the outcome is yes.
What does β₀ mean in Logistic regression?
β₀ (said “beta sub zero, or beta naught”) is the intercept.
Can I listen to the equations in my own textbook like this?
Yes. Upload a PDF, EPUB or text file and Aethel reads it aloud with every equation spoken as maths, the way it reads this one, and highlights each sentence as it goes. You can start free.