The Normal distribution

The Normal distribution

Edexcel International A Level Maths: Revision notes

Studying this course? Keep your progress in one place.

Key facts

  • X∼N(μ,σ2)X\sim\mathrm{N}(\mu,\sigma^2): a symmetrical bell shape with mean μ\mu and variance σ2\sigma^2 (standard deviation σ\sigma).
  • Standardise with Z=X−μσ∼N(0,1)Z=\dfrac{X-\mu}{\sigma}\sim\mathrm{N}(0,1) and use the table of Φ(z)=P(Z<z)\Phi(z)=P(Z<z).
  • Sketch and shade, then use symmetry: P(Z>z)=1−Φ(z)P(Z>z)=1-\Phi(z) and P(Z<−z)=1−Φ(z)P(Z<-z)=1-\Phi(z).
  • To find xx from a probability, find zz from the percentage points, then x=μ+zσx=\mu+z\sigma.
  • With unknown μ\mu and σ\sigma, standardise each condition to form simultaneous equations.

Shape and notation

A Normally distributed variable has a symmetrical bell-shaped density centred on the mean μ\mu, written X∼N(μ,σ2)X\sim\mathrm{N}(\mu,\sigma^2).

A continuous random variable XX that is Normally distributed has the notation X∼N(μ,σ2)X\sim\mathrm{N}(\mu,\sigma^2), where μ\mu is the mean and σ2\sigma^2 is the variance, so the standard deviation is σ\sigma.

The density is a symmetrical bell shape centred on μ\mu, with mean, median and mode all equal to μ\mu.

About 68% of values lie within one standard deviation of the mean and about 95% within two. The total area under the curve is 1, so P(X<μ)=0.5P(X<\mu)=0.5. Because XX is continuous, P(X=x)=0P(X=x)=0.

−6−4−22468100.10.20.30.40.50.60.70.8xyN(m, s²)
Normal curves: change the mean mm to slide the curve and the standard deviation ss to change its width.

X∼N(50,16)X\sim\mathrm{N}(50,16). What is the standard deviation of XX?

The standard Normal variable Z

Subtract the mean and divide by the standard deviation: ZZ counts how many standard deviations XX is above or below the mean.

To use tables, standardise.

ZZ counts how many standard deviations XX is above (z>0z>0) or below (z<0z<0) the mean.

The table of the cumulative distribution function gives Φ(z)=P(Z<z)\Phi(z)=P(Z<z) for positive zz. Interpolation is not needed.

  • StandardiseZ=X−μσ∼N(0,1)Z=\dfrac{X-\mu}{\sigma}\sim\mathrm{N}(0,1)

Worked example

Eggs have mass X∼N(60,42)X\sim\mathrm{N}(60,4^2). Standardise x=66x=66.

X∼N(100,102)X\sim\mathrm{N}(100,10^2). What is zz for x=85x=85?

Finding probabilities with Φ

Sketch the curve and shade the area, then use symmetry to turn each probability into table values Φ(z)\Phi(z).

Always sketch the curve and shade the required area, then use symmetry.

P(Z<z)=Φ(z)P(Z<z)=\Phi(z), and P(a<Z<b)=Φ(b)−Φ(a)P(a<Z<b)=\Phi(b)-\Phi(a).

For a symmetrical interval, P(−z<Z<z)=2Φ(z)−1P(-z<Z<z)=2\Phi(z)-1.

The tables list only positive values, so for a negative zz use Φ(−z)=1−Φ(z)\Phi(-z)=1-\Phi(z).

  • P(Z>z)P(Z>z)1−Φ(z)1-\Phi(z)
  • P(Z<−z)P(Z<-z)1−Φ(z)1-\Phi(z)
  • P(a<Z<b)P(a<Z<b)Φ(b)−Φ(a)\Phi(b)-\Phi(a)

Worked example

T∼N(42,32)T\sim\mathrm{N}(42,3^2). Find P(T>45)P(T>45).

Worked example

Heights N(164,62)\mathrm{N}(164,6^2). Find P(158<X<173)P(158<X<173).

Given Φ(1.2)=0.8849\Phi(1.2)=0.8849, what is P(Z<−1.2)P(Z<-1.2)?

Finding a value from a probability

Find zz with Φ(z)=p\Phi(z)=p from the percentage points table, then use x=μ+zσx=\mu+z\sigma.

To find xx given a probability, work backwards. Use the table of percentage points to find zz with Φ(z)=p\Phi(z)=p, then solve x=μ+zσx=\mu+z\sigma.

For a lower tail with p<0.5p<0.5, find zz for 1−p1-p and make it negative.

Upper 10%

Probability p:
0.90
Value of z:
1.2816

Upper 5%

Probability p:
0.95
Value of z:
1.6449

Upper 2.5%

Probability p:
0.975
Value of z:
1.9600

Upper 1%

Probability p:
0.99
Value of z:
2.3263

Worked example

Find the height exceeded by the tallest 5% of women, N(164,62)\mathrm{N}(164,6^2).

P(Z<z)=0.10P(Z<z)=0.10. Which is correct?

Unknown mean and standard deviation

Standardise each given probability to form simultaneous equations in μ\mu and σ\sigma. One condition is enough if the other parameter is known.

When two probabilities are given and both μ\mu and σ\sigma are unknown, standardise each condition to form simultaneous equations.

If only one condition is given with σ\sigma known (or μ\mu known), a single equation is enough.

Keep zz-values to at least 3 s.f. (use the percentage-point table) and round only at the end.

  • Lower 10% below 995 g995−μσ=−1.2816\dfrac{995-\mu}{\sigma}=-1.2816
  • Upper 5% above 1010 g1010−μσ=1.6449\dfrac{1010-\mu}{\sigma}=1.6449

Worked example

10% of bags are below 995 g and 5% are above 1010 g. Find μ\mu and σ\sigma.

How many equations do you need if both μ\mu and σ\sigma are unknown?

Presenting working

Write each step: the distribution, the standardised zz, the table reading and the answer in context with units.

Write each step: the distribution with its parameters, the standardised zz value, the table reading and the final probability or value.

Give probabilities to 4 decimal places from the tables, or 3 significant figures at the end.

Remember the context: state whether a probability is for a mass, time or height, with units. A small probability does not mean impossible, because the Normal curve extends in both directions.

  1. 1

    State the distribution

    write X∼N(μ,σ2)X\sim\mathrm{N}(\mu,\sigma^2)

  2. 2

    Standardise

    find z=x−μσz=\frac{x-\mu}{\sigma}

  3. 3

    Read the table

    use Φ(z)\Phi(z) and symmetry

  4. 4

    Answer in context

    include units and the right accuracy

A clear Normal distribution answer.

Which step comes straight after writing X∼N(μ,σ2)X\sim\mathrm{N}(\mu,\sigma^2)?

Try an exam question

The masses of apples are modelled by X∼N(150,122)X\sim\mathrm{N}(150,12^2) grams. (a) Find P(X>165)P(X>165). (b) Find the mass that is exceeded by 10% of apples.

[5 marks]

That's the notes covered.

Carry on to the next subtopic.

Try a question

Find the value of Z=X−μσZ=\frac{X-\mu}{\sigma} when X=66X=66.

1 mark, marked against the mark scheme

Revision notes