liveBook · Manning

Bernoulli distribution

As explained in Grokking Bayes

A probability distribution for a single experiment with two possible outcomes, parameterized by the probability of success.

When an uncertain question has only two possible answers, a Bernoulli distribution represents your belief about which answer will occur. For example, whether it will rain today can be encoded as a binary random variable: 1 means rain and 0 means no rain. section 1.5

A Bernoulli distribution has one parameter, p, the probability of success. The probability of 1 is p, while the probability of 0 is 1 − p; together, these probabilities sum to 1. section 1.5

A 30% chance of rain

Represent a 30% chance of rain with p = 0.3. The corresponding probability of no rain is 1 − 0.3 = 0.7. A different weather belief simply uses a different value of p. section 1.5

How Grokking Bayes covers it

How deep it goes
Defines the binary outcomes and parameter p, but stops short of broader distributional calculations.
Examples
Uses a rain-or-no-rain forecast and compares 30% and other beliefs about rain.
What it assumes
Introduces probability distributions and binary random variables through an everyday weather example.
What you can do after
Apply it to encode a two-outcome belief and calculate the complementary outcome probability.
How it connects
Connects probability distributions, binary random variables, and categorical distributions.

Also explained in

Compare how 3 books explain Bernoulli distribution →

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.