Lesson 5 of 8 · 12 min

Bernoulli distributions

Model a single trial with two outcomes by a Bernoulli random variable, and derive and use its mean pp and variance p(1−p)p(1 - p).

What you will be able to do

  • Recognise two-outcome situations and model them with a Bernoulli random variable with parameter pp.
  • Derive and use E⁡(X)=p\E(X) = p and Var⁡(X)=p(1−p)\Var(X) = p(1 - p).
  • Work back from a mean or variance to the value of pp.

A goal kicker lines up a conversion. Either it goes over or it doesn't. A seed germinates or it doesn't. A customer buys the extended warranty or doesn't. Lots of chance situations have exactly two outcomes, and one random variable handles them all.

Definition

Bernoulli random variable

A Bernoulli trial is one chance experiment with exactly two outcomes, called success and failure, where P(success)=pP(\text{success}) = p.

The Bernoulli random variable counts the successes in one trial: X=1X = 1 for a success and X=0X = 0 for a failure.

xx01
P(X=x)P(X = x)1−p1 - ppp

The number pp is the parameter of the distribution.

"Success" is just a label for the outcome you are counting; it doesn't have to be good. In a study of side effects, "success" might be that a patient has a headache.

Interactive

A Bernoulli distribution

Move pp. Where is the mean? For which value of pp is the spread greatest?

Using the general formulas from the last two lessons:

E⁡(X)=0(1−p)+1(p)=p,\E(X) = 0(1 - p) + 1(p) = p, E⁡(X2)=02(1−p)+12(p)=p,soVar⁡(X)=p−p2=p(1−p).\E(X^2) = 0^2(1 - p) + 1^2(p) = p, \quad\text{so}\quad \Var(X) = p - p^2 = p(1 - p).

Key idea

Mean and variance of a Bernoulli random variable

If XX is Bernoulli with parameter pp, then

E⁡(X)=pandVar⁡(X)=p(1−p),σ=p(1−p).\E(X) = p \quad\text{and}\quad \Var(X) = p(1 - p), \qquad \sigma = \sqrt{p(1 - p)}.

A goal kicker converts 85% of attempts. Let X=1X = 1 if the next kick is successful and X=0X = 0 otherwise. Find Var⁡(X)\Var(X).

Which of these can be modelled by a Bernoulli random variable? Select all that apply.

Select all that apply.

Worked example

Find p from a Bernoulli variance

  • Simple familiar
  • Technology free

Let X=1X = 1 if a goal kicker's next conversion attempt is successful and X=0X = 0 otherwise. The kicker succeeds more often than not, and Var⁡(X)=0.16\Var(X) = 0.16.

(a) Determine pp, the probability of success.

(b) Determine the standard deviation of XX.

4 steps

Try it yourself first, then check the working one step at a time.

Method

  1. Set p(1−p)p(1 - p) equal to the variance.

  2. Solve the quadratic; the two solutions add to 1.

  3. Use the context to choose one.

Watch out for

  • Giving both solutions without choosing.

  • Treating 0.16 as the standard deviation.

A Bernoulli random variable XX has Var⁡(X)=0.24\Var(X) = 0.24 and P(X=1)<0.5P(X = 1) < 0.5. Find pp.

Common mistake

Mean, variance or standard deviation?

For a Bernoulli variable the mean is pp, the variance is p(1−p)p(1 - p) and the standard deviation is p(1−p)\sqrt{p(1 - p)}. Three different numbers; read which one the question wants.

Exam tip

Two solutions for p

p(1−p)p(1 - p) is symmetric about p=0.5p = 0.5, so a given variance (below 0.25) comes from two values of pp that add to 1. Use the information in the question to choose, and say why.

Summary

  • A Bernoulli trial has two outcomes, success with probability pp and failure with 1−p1 - p.

  • The Bernoulli random variable is 1 for a success and 0 for a failure.

  • E⁡(X)=p\E(X) = p and Var⁡(X)=p(1−p)\Var(X) = p(1 - p), which is largest, 0.25, when p=0.5p = 0.5.

  • A given variance usually comes from two values of pp that add to 1.

Finished the lesson?

Try the 3 quick checks above and the lesson completes itself, or mark it complete now.

Lesson 5 of 80% read

Save your progress with a free account