It models a single trial of a random experiment that has exactly two possible outcomes, often labeled as "success" (usually encoded as 1) and "failure" (usually encoded as 0).
Think of it as a single coin flip (potentially biased).
A Random Variable\(X\) follows a Bernoulli distribution if it takes the value \(1\) (success) with probability \(p\), and the value \(0\) (failure) with probability \(1-p\).
Notation: \(X \sim \text{Bernoulli}(p)\) (read as "X follows a Bernoulli distribution with parameter p").
The Bernoulli distribution is the fundamental building block for several other more complex discrete distributions:
Binomial Distribution: Represents the number of successes in a fixed number (\(n\)) of independent Bernoulli trials.
[[08_Geometric_Distribution|Geometric Distribution]]: Represents the number of Bernoulli trials needed to get the first success.
[[09_Negative_Binomial_Distribution|Negative Binomial Distribution]]: Represents the number of Bernoulli trials needed to get a fixed number (\(r\)) of successes.
Used in Logistic Regression where the outcome for each data point is often modeled as a Bernoulli variable (e.g., predicting click/no-click).
Classic Texts: (e.g., "Probability and Statistics for Engineers and Scientists" by Walpole, Myers, Ye; "A First Course in Probability" by Sheldon Ross) - Consult relevant chapters.