Bernoulli Mean and Variance
What you'll learn: You'll discover the formulas for the expected value and variance of a Bernoulli random variable and understand what they tell you about coin-flip-like experiments.
Understanding the Formulas
You already know that a Bernoulli random variable X represents a single trial with two outcomes: success (X=1) with probability p, or failure (X=0) with probability 1-p.
Now let's find the expected value (mean):
E(X) = 0·(1-p) + 1·p = p
That's it! The expected value of a Bernoulli trial equals the probability of success. If you flip a coin with p=0.7 for heads, the "average outcome" is 0.7. This doesn't mean you'll ever observe 0.7—it means over many flips, the average approaches 0.7.
For variance, recall that Var(X) = E(X²) - [E(X)]².
First, find E(X²):
- When X=0: 0² = 0
- When X=1: 1² = 1
So E(X²) = 0·(1-p) + 1·p = p
Therefore: Var(X) = p - p² = p(1-p)
What Does This Mean?
The variance p(1-p) measures uncertainty. Notice:
- When p=0.5 (fair coin), variance = 0.25 (maximum uncertainty)
- When p approaches 0 or 1 (almost certain outcome), variance approaches 0 (minimal uncertainty)
- The standard deviation is √[p(1-p)]
Key Takeaway: For a Bernoulli random variable, the mean E(X)=p represents the probability of success, while the variance Var(X)=p(1-p) quantifies the uncertainty, reaching maximum at p=0.5 and minimum at the extremes.