Poisson Approximation to the Binomial Explained with Worked Examples

When is large and is small, the binomial distribution is well approximated by .

Write 2x or 2*x to multiply, ^ for powers, and ; between simultaneous equations. You can also type factor, expand, derivative of, integrate … from a to b, limit … as x->0 or 15% of 80.

Try an example

Before you start

This page is easier if you are comfortable with:

What Poisson approximation to the binomial means

For , . With fixed and , and , giving the Poisson probability. The binomial itself is in the binomial distribution and the coefficients in binomial coefficients.

A common rule of thumb is and (or ). For , the Poisson gives against the exact 0.1823. The means match (), and the binomial variance is close to the Poisson variance when is small; see the mean and variance of a binomial.

The approximation was important before computers because the Poisson probabilities are easier to calculate. For moderate and large , the normal curve fits better: the normal approximation to the binomial. A hypothesis test with rare events uses a binomial proportion test.

What is it used for?

For many trials with a small success probability, the binomial and Poisson probabilities are almost the same, and the Poisson is easier to calculate.

How to use poisson approximation to the binomial

  1. Check that is large and is small.
  2. Set .
  3. Use the Poisson probability instead of the binomial.
  4. Compare with the exact value if possible.

Graph

B(100, 0.02) and Po(2) probabilities for r = 0 to 7r = 3: 0.182 and 0.1800123456700.050.10.150.20.250.3B(100, 0.02) and Po(2) probabilities for r = 0 to 7rprobabilityGraph with x from -0.5 to 7.5 and y from 0 to 0.32. Shows binomial B(100, 0.02) (8); Poisson Po(2) (8); point r = 3: 0.182 and 0.180 at (3, 0.1823).

The two sets of points are almost on top of each other.

Worked examples

These examples are worked by hand, step by step, so you can compare your own working.

Example 1: B(100, 0.02)

Approximate for .

Worked example
Answer
0.1804 (exact 0.1823)0.1804\ (\text{exact }0.1823)
  1. λ = np
  2. Poisson
  3. Exact binomial

Example 2: B(200, 0.01)

Approximate for .

Worked example
Answer
0.1353 (exact 0.1340)0.1353\ (\text{exact }0.1340)
  1. λ
  2. Poisson
  3. Exact

Example 3: B(50, 0.04)

Approximate for .

Worked example
Answer
0.4060 (exact 0.4005)0.4060\ (\text{exact }0.4005)
  1. λ
  2. Poisson
  3. Exact

Special cases

A few situations come up again and again. Knowing them saves time and helps you spot answers that cannot be right.

B(100, 0.02)

λ = np

2.

Po(2)

P(X = 1)

0.2707.

B(100, 0.02) exactly

P(X = 1)

0.2707.

B(200, 0.01)

λ

2.

Word problems

For large n and small p, B(n, p) ≈ Po(np). A good rule is n ≥ 50 and np ≤ 5 or so.

λ

n = 100, p = 0.02.

  1. 100 × 0.02
  2. Check: ✓2

Poisson P(1)

λ = 2.

  1. 2e^(−2)
  2. Check: ✓0.2707

Exact P(1) Higher

B(100, 0.02).

  1. 100(0.02)(0.98)⁹⁹
  2. Check: close ✓0.2707

Another Higher

n = 200, p = 0.01.

  1. 200 × 0.01
  2. Check: ✓2

Common mistakes

Use for small n

small n ✗

large n, small p ✓

Conditions.

λ = n/p

n/p ✗

np ✓

Mean.

Use when p is large

p = 0.5 ✗

small p only ✓

Rare events.

Exact equality

equal ✗

approximately ✓

Approximation.

Exam tips

  1. λ = np.
  2. n large, p small.
  3. Compare with the exact value.
  4. Check np ≤ 5.
  5. If p near 1, swap success and failure.

Practice questions

Try each one on paper, then open the answer. Use the solver above to see the full working.

1. λ for B(500, 0.004) Foundation

2

2. λ for B(80, 0.05) Foundation

4

3. Po P(0) for λ = 2, 4 dp Foundation

0.1353

4. Po P(2) for λ = 1, 4 dp Higher

0.1839

5. λ for B(1000, 0.003) Higher

3

6. Po P(1) for λ = 3, 4 dp Higher

0.1494

Practice generator

Unlimited fresh questions with instant marking. Choose a level, work it out on paper, then type your answer.

Solve: …

Key terms

Poisson approximation
Po(np).
Binomial
B(n, p).
Rare event
Small p.
Mean (np)
λ.
Approximation
Close not exact.
Large n
Many trials.

Frequently asked questions

When can I use the Poisson approximation?

When n is large and p is small.

What is λ?

np.

How good is it?

Close for np small.

What if p is near 1?

Use failures instead.

Which page covers the normal approximation?

normal approximation to the binomial.

Which page covers the binomial?

binomial distribution.

More in Advanced statistics and stochastic processes and the formula sheets.