The normal distribution

The normal distribution

Spec 4.2 · Edexcel A-Level Maths: Revision notes

Studying this course? Keep your progress in one place.

Key facts

  • X∼N(μ,σ2)X\sim N(\mu,\sigma^2): mean μ\mu, variance σ2\sigma^2, standard deviation σ\sigma. The curve is symmetrical about μ\mu.
  • Probability is area; P(X=a)=0P(X=a)=0 and P(X<μ)=0.5P(X<\mu)=0.5.
  • Points of inflection are at μ±σ\mu\pm\sigma; larger σ\sigma gives a wider, flatter curve.
  • Standardise: Z=X−μσZ=\dfrac{X-\mu}{\sigma} with Z∼N(0,12)Z\sim N(0,1^2).
  • With two unknowns, convert two probabilities to zz-values and solve simultaneously.

The Normal model

A continuous, symmetrical, bell-shaped model N(μ,σ2)N(\mu,\sigma^2) for heights, masses and times.

The Normal distribution is written X∼N(μ,σ2)X\sim N(\mu,\sigma^2). The curve is symmetrical about x=μx=\mu, so mean, median and mode are equal. The total area is 1 and probability is area, so P(X=a)=0P(X=a)=0 and P(X<a)=P(X≤a)P(X<a)=P(X\le a).

A roughly symmetrical, bell-shaped histogram suggests a Normal model, with its mean and standard deviation estimating μ\mu and σ\sigma. The formula for the density is not needed.

4060801001201401600.020.040.060.08xyN(μ, σ²)
The Normal curve N(μ, σ²) is symmetrical and bell-shaped. Change μ to slide it and σ to make it taller and narrower or lower and wider.
  • X∼N(μ,σ2)X\sim N(\mu,\sigma^2)mean μ\mu, variance σ2\sigma^2
  • N(100,152)N(100,15^2)standard deviation 1515

X∼N(50,16)X\sim N(50,16). What is the standard deviation?

Shape and points of inflection

A larger σ\sigma gives a wider, flatter curve; changing μ\mu slides it sideways.

The curve peaks at x=μx=\mu and the points of inflection are at x=μ±σx=\mu\pm\sigma. The area stays 1, so a narrower curve is taller.

For H∼N(42,32)H\sim N(42,3^2) the maximum is at h=42h=42 and the points of inflection are at h=39h=39 and h=45h=45.

−10−8−6−4−22468100.10.20.30.40.50.60.70.8xyNormal density
Normal curve N(m, s²): m moves it, s changes the width.

H∼N(42,32)H\sim N(42,3^2). Where are the points of inflection?

Finding probabilities

Use the Normal cumulative distribution function with μ\mu and σ\sigma, after sketching the area.

Enter σ\sigma, not σ2\sigma^2. P(a<X<b)P(a<X<b) uses both limits; P(X>a)=1−P(X<a)P(X>a)=1-P(X<a). By symmetry, P(X<μ−k)=P(X>μ+k)P(X<\mu-k)=P(X>\mu+k).

For H∼N(42,32)H\sim N(42,3^2): P(38<H<47)=0.861P(38<H<47)=0.861 and P(H>45)=0.1587P(H>45)=0.1587. Sketch and shade the area first so you know whether the answer should be large or small.

303234363840424446485052540.020.040.060.080.10.120.14xypeak at 42(45, 0.081)Density of H
H ~ N(42, 3²) with P(H > 45) shaded.

X∼N(10,22)X\sim N(10,2^2). What is P(X<10)P(X<10)?

Standardising and inverse problems

Convert to Z=X−μσZ=\frac{X-\mu}{\sigma}, which counts the standard deviations from the mean.

To standardise, use Z=X−μσZ=\frac{X-\mu}{\sigma}, where Z∼N(0,12)Z\sim N(0,1^2) is the standard Normal. A zz-value tells you how many standard deviations xx is from the mean: for N(100,152)N(100,15^2), 130 has z=2z=2.

For an inverse problem use the inverse Normal function. If P(Z<z)=0.9P(Z<z)=0.9 then z=1.2816z=1.2816; 0.9750.975 gives z=1.96z=1.96.

−4−3−2−112340.10.20.30.4xyz = 2N(0, 1²)
The standard Normal curve N(0, 1²). For N(100, 15²), x = 130 gives z = 2, and the shaded area is P(Z < 2) = 0.9772.
  • StandardiseZ=X−μσZ=\dfrac{X-\mu}{\sigma}

Worked example

X∼N(100,152)X\sim N(100,15^2). Find the score exceeded by 10% of people.

X∼N(100,152)X\sim N(100,15^2). What is zz for x=130x=130?

Finding unknown μ\mu and σ\sigma

With one unknown use one equation; with two, convert two probabilities to zz-values and solve simultaneously.

If one of μ\mu, σ\sigma is unknown, standardise using a given probability and solve. If both are unknown you are given two probabilities: convert each to a zz-value and solve the simultaneous equations.

  1. 1

    Convert

    Turn each probability into a z-value.

  2. 2

    Standardise

    Write (x − μ)/σ = z for each.

  3. 3

    Solve

    Solve the two equations simultaneously.

  4. 4

    Check

    Lower tails give negative z.

Two unknown parameters

Worked example

L∼N(μ,σ2)L\sim N(\mu,\sigma^2) with P(L<200)=0.1P(L<200)=0.1 and P(L>260)=0.2P(L>260)=0.2. Find μ\mu and σ\sigma.

If P(X<a)=0.3P(X<a)=0.3, is the zz-value for aa positive or negative?

Try an exam question

M∼N(60,42)M\sim N(60,4^2). Find P(M>64)P(M>64) and the value mm such that P(M<m)=0.9P(M<m)=0.9.

[4 marks]

That's the notes covered.

Carry on to the next subtopic.

Try a question

Find P(H>45)P(H>45).

1 mark, marked against the mark scheme

Revision notes