Bernoulli Distribution: PMF, Expectation, and Variance
What you'll learn: You'll derive the exact formulas for the probability mass function, mean, and variance of a Bernoulli random variable.
The Core Idea
A Bernoulli random variable X with parameter p models a single success/failure trial. It takes value 1 (success) with probability p and value 0 (failure) with probability 1 - p. Now let's derive its key properties formally.
Probability Mass Function (PMF)
The PMF assigns probabilities to the only two possible outcomes:
- P(X = 1) = p
- P(X = 0) = 1 - p
We can write this compactly as: P(X = x) = p^x · (1 - p)^(1-x) for x ∈ {0, 1}
This clever formula works because when x = 1, we get p¹·(1-p)⁰ = p, and when x = 0, we get p⁰·(1-p)¹ = 1 - p.
Expectation (Mean)
Using the definition of expectation for discrete random variables:
E[X] = Σ x · P(X = x) = 0·(1 - p) + 1·p = p
The average outcome is simply the success probability—intuitive when you think about long-run frequencies.
Variance
Using the shortcut formula Var(X) = E[X²] - (E[X])²:
First, E[X²] = 0²·(1 - p) + 1²·p = p
Then: Var(X) = p - p² = p(1 - p) = pq (where q = 1 - p)
The variance is maximized when p = 0.5 (maximum uncertainty) and minimized at p = 0 or p = 1 (certainty).
Key Takeaway: For a Bernoulli(p) random variable: PMF is P(X = x) = p^x(1-p)^(1-x), E[X] = p, and Var(X) = p(1-p).