Confidence intervals for a Normal mean

Confidence intervals for a Normal mean

Spec 3.3–4 · Edexcel International A Level Further Maths: Revision notes

Studying this course? Keep your progress in one place.

Key facts

  • A confidence interval is a range likely to contain μ\mu; about 95%95\% of intervals built the same way contain it.
  • With σ\sigma known the limits are xˉ±zσn\bar x\pm z\frac{\sigma}{\sqrt n}; z=1.645,1.96,2.326,2.576z=1.645, 1.96, 2.326, 2.576 for 90%,95%,98%,99%90\%, 95\%, 98\%, 99\%.
  • Higher confidence or larger σ\sigma widens the interval; larger nn narrows it.
  • Round a required sample size up.
  • A c%c\% interval matches a two-tailed test at the (100−c)%(100-c)\% level.

What an interval is

It is a range of plausible values for the parameter, built from a sample.

A confidence interval is a range calculated from a sample that is likely to contain the unknown parameter. The confidence level (for example 95%95\%) says how reliable the method is: if you built an interval from each of many random samples, about 95%95\% would contain the true parameter.

μ\mu is fixed; it is the interval that changes from sample to sample. A higher level gives a wider interval, and a larger sample a narrower one.

−4−3−2−112340.10.20.30.40.5xy−1.961.96Distribution of x̄ (standardised)
Sample means are Normally distributed about μ. The central 95% lie within 1.96 standard errors of μ, which is why a 95% interval is x̄ ± 1.96σ/√n.
  • 95% confidenceabout 95% of intervals contain μ

A 95% confidence interval for a mean is (49.2, 54.8). Which statement is correct?

Limits with known variance

Use xˉ±zσn\bar x\pm z\frac{\sigma}{\sqrt n}, where zz cuts off the central area.

If X∼N(μ,σ2)X\sim N(\mu,\sigma^2) with σ\sigma known, then Xˉ∼N(μ,σ2n)\bar X\sim N\left(\mu,\frac{\sigma^2}{n}\right). The confidence limits are xˉ±zσn\bar x\pm z\frac{\sigma}{\sqrt n}, where σn\frac{\sigma}{\sqrt n} is the standard error and zz cuts off the required central area:

  • 90%90\%: z=1.645z=1.645
  • 95%95\%: z=1.96z=1.96
  • 98%98\%: z=2.326z=2.326
  • 99%99\%: z=2.576z=2.576
−4−3−2−112340.10.20.30.4xy-1.961.96N(0, 1)
The central 95%95\% of N(0,1)N(0,1) lies between z=−1.96z=-1.96 and z=1.96z=1.96.

Worked example

σ=8\sigma=8, n=25n=25, xˉ=1012\bar x=1012. Find a 95%95\% confidence interval for μ\mu.

Which zz-value gives a 99% confidence interval?

Width and sample size

The width is 2zσn2z\frac{\sigma}{\sqrt n}; solve for nn and round up.

The interval is symmetrical about xˉ\bar x, with half-width zσnz\frac{\sigma}{\sqrt n}. The width falls if nn increases, and rises if the confidence level or σ\sigma rises.

For a required width ww: n≥(2zσw)2n\ge\left(\frac{2z\sigma}{w}\right)^2, and always round nn up. Working backwards: a 95%95\% interval (49.2,54.8)(49.2,54.8) with σ=10\sigma=10 has xˉ=52\bar x=52 and half-width 2.8=1.96×10n2.8=1.96\times\frac{10}{\sqrt n}, so n=49n=49.

102030405060708090100510152025303540Sample size nWidthn = 49: 5.6n = 4: 19.6w = 2 × 1.96 × 10/√n
Width of a 95% interval for a Normal mean with σ = 10: w = 39.2/√n. Quadrupling n halves the width.
  • Width2zσn2z\dfrac{\sigma}{\sqrt n}
  • Sample sizen≥(2zσw)2n\ge\left(\dfrac{2z\sigma}{w}\right)^2

Worked example

σ=4.5\sigma=4.5. Find the smallest nn so that a 95%95\% interval has width at most 22.

Doubling the sample size makes the width of the interval:

Interpreting an interval

It is a set of plausible values; the confidence level describes the method, not one interval.

A confidence interval does not say that 95%95\% of the population values lie in it, and it does not give the probability that a fixed μ\mu is in a particular interval. Different samples give different intervals, and even a correct method fails to capture μ\mu about 5%5\% of the time at the 95%95\% level.

In context, say "we are 95%95\% confident that the mean journey time lies between 50.550.5 and 54.354.3 minutes". The interval is a set of plausible values.

4849505152535455565750.5x̄ = 52.454.3

Valid

  • Repeated intervals: about 95% contain μ
  • The interval is a set of plausible values

Not valid

  • 95% of the population lies in it
  • Probability 0.95 that μ is in this interval

Interval A is (50.5, 54.3) at 95%. A 99% interval from the same data would be:

Try an exam question

The mass of a bag of flour is Normal with standard deviation 1010 g. A random sample of 4949 bags has mean mass 5252 g above the nominal value. (a) Find a 95%95\% confidence interval for the mean excess. (b) Comment on a claim that the mean excess is 5656 g.

[4 marks]

That's the notes covered.

Carry on to the next subtopic.

Try a question

What is the standard error of the sample mean?

1 mark, marked against the mark scheme

Revision notes