University of Sydney · FACULTY OF STATISTICS

QBUS5001 Chap.7 Estimation and Confidence Intervals

- one subject, every graph, every model, every mark
6 Chapters4-page Bible
Our own words - no uploaded lecturer files
Updated for this semester
Chapter 7 of 10 · QBUS5001

Estimation and Confidence Intervals

This chapter builds confidence intervals as a point estimate plus or minus a critical value times standard error. It covers means, proportions and sample-size planning, with attention to Normal versus Student t choices. Interpretation stays with the population parameter and the long-run procedure.

Precision improves with sample size, but no narrow interval can compensate for biased collection or dependence ignored by the standard error. The margin of error separates the observed estimate from each endpoint and combines a critical value with sampling variability. Higher confidence uses a larger critical value and therefore a wider interval when the data are fixed.

Larger samples reduce standard error through a square-root relationship, so halving the margin of error usually requires far more than doubling the sample. Mean intervals choose between a known-spread Normal structure and an estimated-spread Student t structure. Proportion intervals use a different standard error based on the estimated share and require suitable success and failure counts.

Sample-size planning works backward from a desired margin of error and must round upward. An interval describes a long-run method that captures the fixed parameter at the advertised rate, not a changing posterior probability about this one completed interval.

In this chapter

What this chapter covers

  • 01

    Point estimates and margins of error

  • 02

    Confidence as long-run procedure performance

  • 03

    Mean intervals with known or estimated variability

  • 04

    Proportion intervals and count conditions

  • 05

    Confidence level, sample size and interval width

  • 06

    Sample-size determination for a target margin

Worked example · free

Construct an interval for a population mean

Q [5 marks]. A random sample of 64 orders has mean fulfilment time 38.2 minutes and sample standard deviation 8.0 minutes. Build an approximate 95% interval. The mark allocation used here is not a University assessment scheme.
Repeated confidence intervals target a fixed parameterNine horizontal intervals cross a vertical population-mean line while one misses it. fixed mean miss
The vertical parameter does not move; repeated random samples move the interval endpoints, and the confidence level describes the long-run capture rate of this procedure.
  • +1Use sample mean 38.2 as the point estimate.
  • +1Estimate standard error as 8 divided by square root 64 equals 1 minute.
  • +1Use the approximate 95% critical value 1.96 and margin 1.96 minutes.
  • +1Calculate endpoints 36.24 and 40.16 minutes.
  • +1Interpret the interval for the population mean represented by the sampling process.
The approximate 95% confidence interval for the population mean is 36.24 to 40.16 minutes.
Sia tip — An interval for a mean does not describe the range containing most individual observations.
Glossary

Key terms

Point Estimate
A single sample-based value used to estimate a population parameter.
Margin of Error
The critical value multiplied by standard error in a confidence interval.
Confidence Level
The long-run proportion of intervals from a procedure that capture the fixed parameter.
Critical Value
A distribution cutoff chosen to produce the desired confidence level or test size.
Sample Size
The number of sampled units, which affects standard error through a square-root relationship.
FAQ

Estimation and Confidence Intervals FAQ

What does 95% confidence mean?

It describes the procedure: across many comparable random samples, about 95% of intervals built this way would capture the fixed population parameter. It does not mean this already calculated interval has a 95% probability of containing a moving parameter.

Why does higher confidence make an interval wider?

Higher confidence requires capturing the fixed parameter across a larger share of repeated samples, so the procedure uses a more distant critical value. With the same estimate and standard error, that larger multiplier increases the margin of error.

How do I reduce the margin of error?

Increase sample size, reduce measurement variability through better design or accept a lower confidence level if appropriate. Because standard error changes with the square root of sample size, halving a margin usually requires roughly four times as many observations.

Study strategy

Exam move

Memorise the structure rather than isolated formulas: estimate, critical value, standard error, endpoints and contextual interpretation. Compare a mean interval with a proportion interval to see where the standard error changes. Recalculate one interval at a higher confidence level and larger sample size, predicting the direction before doing arithmetic.

Round designed sample sizes upward and discuss the collection mechanism separately. Build intervals in five labelled lines: point estimate, standard error, critical value, margin of error and endpoints. This layout makes it easy to locate an arithmetic or model-choice error. Before recomputing, predict how the width changes when confidence, variability or sample size changes.

Contrast a known population spread with an estimated sample spread so the Normal versus Student t decision is tied to information, not habit. For proportions, check that the estimated counts supporting the approximation are adequate. Rewrite every interpretation until it names the population parameter and the repeated-sampling procedure. Then add one design caveat about random selection or dependence.

That final caveat prevents a precise numerical interval from being presented as stronger evidence than the collection process can support. Audit the endpoints against the point estimate. For a symmetric textbook interval, the estimate should sit midway between them and each endpoint should be one margin of error away.

Then compare the interval with any operational threshold without claiming that crossing the threshold proves equality. The interval describes estimation uncertainty and should be read alongside practical decision tolerances.

Working through Estimation and Confidence Intervals in QBUS5001? Sia is AskSia’s AI Statistics tutor — ask any QBUS5001 Estimation and Confidence Intervals question and get a clear, step-by-step explanation grounded in how QBUS5001 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.

A+Everything unlocked
Unlocks this Bible + all 202 of your subjects at University of Sydney, plus 1,000+ more Bibles across the AskSia library.
Sia - your QBUS5001 tutor, unlimited, worked the way the exam marks it
The full 4-page Bible + practice bank with worked solutions
Chrome extension - sync your LMS so Sia knows your deadlines
Bilingual EN / Chinese on every Bible and every Sia answer
$0.99 Trial
30-day money-back · cancel in one tap · how it works
QBUS5001 · Foundation in Data Analytics for Business - independent study guide on the AskSia Library. More University of Sydney subjects · Microeconomics across all universities