How accurate are weather forecasts?: a maths EPQ idea
A title to start from
How accurate are UK weather forecasts, and what is the fairest way to measure accuracy?
Why it works as an EPQ
You collect forecasts and outcomes yourself and choose between scoring rules, which is a genuine judgement.
Scope and difficulty
Solid. Solid. Six to eight weeks of daily data for one place.
The maths
Builds on these A Level topics: Sampling and data · Probability · Hypothesis testing.
You would learn:
- Scoring rules such as the Brier score
- Calibration of probability forecasts
- Skill relative to a simple baseline
One possible plan
- Record forecasts for one location and the outcomes each day.
- Score them in two or three ways.
- Compare with a baseline such as 'tomorrow will be like today'.
- Judge which measure is fairest and how good the forecasts are.
Pitfalls
- Gaps in daily data collection.
- Treating '40% chance of rain' as wrong whenever it rains.
Where to start reading
- Search for: Brier score probability forecast verification
- Met Office forecast and observation pages
Similar ideas
- Is the lottery a fair game?Is a national lottery a fair game, and what is a ticket really worth?
- Benford's law in UK dataDoes Benford's law hold for UK data sets, and could it detect made-up numbers?
- Shared birthdays in real classesHow likely are shared birthdays in real classes, given that births are not spread evenly through the year?
- Why do opinion polls miss?Why do opinion polls sometimes get UK general elections wrong?
All statistics and data ideas · all 93 ideas
Your EPQ must be your own work. These pages coach: ideas, structure, checklists and planning. Submitting text, proofs, code or analysis written by someone else or by an AI tool as your own is malpractice. How to use help and AI honestly.