Normal approximation to the binomial and Poisson
Studying this course? Keep your progress in one place.
Key facts
- Use when is large and is close to (, ).
- Use when is large (usually ).
- The second parameter of is the variance.
- Always apply a continuity correction: move the boundary outwards if the integer is included.
When it is appropriate
When the discrete distribution is large and roughly symmetrical.
Binomial and Poisson probabilities are tedious to add up when or is large. A normal approximation replaces the discrete distribution with a continuous one with the same mean and variance.
If is very small or very large, the distribution is skewed and the approximation is poor.
| Binomial B(n, p) | Poisson Po(λ) | |
|---|---|---|
| Good when | n large, p close to 0.5 | λ large |
| Usual check | np > 5 and nq > 5 | λ > 10 |
Binomial B(n, p)
- Good when:
- n large, p close to 0.5
- Usual check:
- np > 5 and nq > 5
Poisson Po(λ)
- Good when:
- λ large
- Usual check:
- λ > 10
Which distribution is best approximated by a normal distribution?
Approximating the binomial
, where .
If with large and near , the mean is and the variance is . The second parameter of is the variance, so standardise with the standard deviation .
For : , , so .
- B(n, p)
Worked example
. Write down the normal approximation.
- 1
Mean: .
- 2
Variance: .
. Which is the normal approximation?
Approximating the Poisson
, since the mean and variance are both .
If with large, the Poisson distribution has mean and variance both equal to , so the standard deviation is .
For : with .
- Po(λ)
. What is the normal approximation?
The continuity correction
takes whole numbers but the normal variable is continuous, so each integer becomes the block .
The adjustment is the continuity correction. Rewrite strict inequalities as inclusive ones first: means , so the boundary is .
Ask whether the integer is included. If it is, move the boundary outwards by . If it is not, move it inwards.
Write with a continuity correction.
Worked examples
Approximate, apply the correction, standardise, then read the table.
State the approximating distribution. Apply the continuity correction. Standardise with and use .
For an inverse problem, solve and round up to an integer.
Worked example
. Find .
- 1
Approximate by , so .
- 2
.
- 3
.
- 4
.
Worked example
. Find .
- 1
, so .
- 2
means , so use .
- 3
and .
- 4
Probability .
Worked example
. Find the smallest with .
- 1
, .
- 2
Solve .
- 3
, so .
- 4
Round up to an integer.
. Find the value used to approximate .
Try an exam question
. Use a normal approximation to find .
[5 marks]
- [1] (mean , variance ).
- [1]Continuity correction: .
- [1].
- [1].
- [1].
That's the notes covered.
Carry on to the next subtopic.
Try a question
Which normal distribution approximates ?
1 mark, marked against the mark scheme
Which normal distribution approximates ?
1 mark, marked against the mark scheme