University College London · FACULTY OF STATISTICS

BENV0139 Chap.3 Probability, Normal Models and the CLT

- one subject, every graph, every model, every mark
7 Chapters3-page Bible
Our own words - no uploaded lecturer files
Updated for this semester
Chapter 3 of 8 · BENV0139

Probability, Normal Models and the CLT

Probability, Normal Models and the CLT turns thresholds into areas under stated distributions. A normal model is defined by its mean and standard deviation, and a z score expresses a value's signed distance from the mean in standard-deviation units. The wording of the question selects an upper tail, lower tail, two tails or interval. Sketching that area before using a table or software prevents direction errors.

The Central Limit Theorem concerns suitable sums and sample means under conditions including independence; it does not force a raw histogram to become normal merely because many rows were collected. Daily cycles, grouped observations, heavy tails and dominant influences can break a convenient approximation. Binomial models address counts of successes under fixed, independent and equal-probability trials.

The chapter keeps the target distribution visible so a precise calculation does not answer the wrong probability question.

In this chapter

What this chapter covers

  • 01

    Normal distribution

  • 02

    Standard score

  • 03

    One-sided tails

  • 04

    Two-sided areas

  • 05

    Binomial counts

  • 06

    Central Limit Theorem

  • 07

    Independence, skew, heavy tails and dominant cycles as boundaries on probability models

Worked example · free

Calculate an upper-tail probability

Q [7 marks]. Daily peak temperature is modelled as normal with mean 19.0°C and SD 2.0°C. Find the probability of exceeding 22.0°C. The marks are AskSia practice weighting, not a published UCL marking scheme.
The normal curve marks a threshold at z equals 1.50 and shades the upper-tail probability beyond that standardised value.The normal curve marks a threshold at z equals 1.50 and shades the upper-tail probability beyond that standardised value. z scorez = 1.50upper tailmean
The normal curve marks a threshold at z equals 1.50 and shades the upper-tail probability beyond that standardised value.
  • 1Sketch the normal curve and shade the upper tail beyond 22.0°C. The marks are original practice weighting.
  • 2Calculate z = (22.0 - 19.0) / 2.0 = 1.50.
  • 2Read P(Z > 1.50) as approximately 0.0668.
  • 2Interpret about 6.7% of comparable days under a stable normal model and state the seasonality limitation.
The upper-tail probability is about 0.0668 under the model. Practice marks are not an official UCL allocation.
Sia tip — Shade first and calculate second; the picture distinguishes the requested tail from its complement.
Glossary

Key terms

Standard score
A value's signed distance from a mean measured in standard-deviation units.
Sampling distribution
The distribution of a statistic across repeated samples from the same process.
Binomial model
A probability model for the number of successes in fixed independent trials with a common success probability.
FAQ

Probability, Normal Models and the CLT FAQ

What does a z score tell me?

It tells you how many standard deviations a value lies above or below a mean under the stated scale. It is not itself a probability. Convert the threshold first, then use the question wording to select the correct tail or interval area.

Does a large sample make raw data normal?

No. The CLT concerns the sampling distribution of suitable sums or means, not the shape of individual measurements. Dependence, strong skew, heavy tails and dominant daily or annual cycles can require larger samples or a different model.

When is a binomial model suitable?

Use it for a fixed number of trials that each have two defined outcomes, a common success probability and sufficient independence. Clustered objects, changing detection rates or more than two outcome categories need a revised model or explicit qualification.

Study strategy

Exam move

Draw before computing. For every probability prompt, sketch the distribution, mark the mean and threshold, shade the requested region and estimate whether the answer should be small or large. Then standardise and use a table or software. Practise pairs that differ only in words such as above, below, between and different from; explain why the area changes.

Keep a target label beside each spread: sigma for individual observations and standard error for a sample mean. Build a CLT checklist covering the statistic, independence, sample size, skew, tails and dominant structure. Use environmental time series to practise explaining why a large row count can retain daily dependence.

For binomial questions, verify fixed trial count, binary outcome and common probability before calculating. End each solution with a sentence that names the model and population. If the answer would change across seasons, materials or control regimes, state that boundary rather than reporting a probability as universal. Build a tail-conversion set from one normal model.

Solve above, below, between two values and outside a symmetric interval, drawing a separate shaded curve for each. Before consulting a table, predict whether the area exceeds one half and whether symmetry can reduce the work. After computing, reverse the standardisation to confirm the threshold and translate the probability into expected days only when independence and a stable period make that interpretation sensible.

Create a second card contrasting an individual reading with a mean of sixteen readings; label standard deviation and standard error and explain why they answer different targets. For the CLT, classify examples by independent units rather than rows, especially hourly measurements within days or spectra within fragments. Sketch how daily cycles or heavy tails can slow an approximation.

For a binomial count, challenge each assumption by introducing clustered frames, unequal detection or a changing trial count. Finish every answer with model, population, event, numeric probability and one condition that could invalidate transfer to another material, season or control regime.

Working through Probability, Normal Models and the CLT in BENV0139? Sia is AskSia’s AI Statistics tutor — ask any BENV0139 Probability, Normal Models and the CLT question and get a clear, step-by-step explanation grounded in how BENV0139 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.

A+Everything unlocked
Unlocks this Bible + all 32 of your subjects at University College London, plus 1,000+ more Bibles across the AskSia library.
Sia - your BENV0139 tutor, unlimited, worked the way the exam marks it
The full 3-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