liveBook · Manning

Probability mass function

As explained in Grokking Bayes

A function that assigns the exact probability mass to each outcome of a discrete random variable.

A probability mass function (PMF) assigns the exact probability mass to each outcome of a discrete random variable. For finitely many hypotheses, adding their probability masses gives the total probability, which must equal 1. section 3.2

The distinction from a probability density function matters: a PMF gives an individual outcome’s exact probability, while a PDF describes continuous values through density and requires integration over a range. section 3.2

Discrete versus continuous probability

For finitely many hypotheses, sum the PMF values to obtain probabilities. For a uniform continuous distribution from 0 to 1, the density can equal 1 everywhere, but only areas over ranges represent probabilities. section 3.2

How Grokking Bayes covers it

How deep it goes
Explains exact mass, normalization, and the PMF-PDF distinction, but stops short of computational implementation.
Examples
Uses finitely many hypotheses and a uniform continuous distribution between 0 and 1 to contrast mass with density.
What you can do after
Evaluate discrete uncertainty by identifying exact outcome probabilities and summing probability masses to obtain totals.
How it connects
Connects probability mass functions with probability density functions, discrete distributions, continuous distributions, and integration.

Also explained in

Compare how 4 books explain probability mass function →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Grokking Bayes. The section links point to the parts of the book it draws on.