Learn
Forecasting and scenario analysis are not the same
What is the difference between a forecast and a scenario?
Short answer
A forecast is generated by a model from defined data and assumptions. A scenario begins with an explicit assumption: one a person states, or one that comes from a predefined stress case where a tool offers them. It reports what that assumption would mean, and says nothing about how likely the assumption is.
- Author:
- Neo-Invest.AI
- Reviewer:
- Erman Peker, Founder
- Drafted:
- 2026-08-01
- Published:
- 2026-08-01
- Last reviewed:
- 2026-08-01
- Version:
- 1.0
Key takeaways
- A forecast is produced by a model from data; a scenario starts from an assumption someone states explicitly.
- A scenario result is conditional on its assumption and carries no probability.
- Mechanical scenarios can be deterministic when holdings, prices, rates, assumptions and timestamps are defined.
- Model-based stress scenarios rest on modelling choices and should be read as estimates.
Where each one starts
The difference is in where the premise comes from. A forecast is generated by a model from defined data and assumptions: it reads conditions and produces an estimate. A scenario starts from an assumption someone states — equities fall by a stated percentage, an exchange rate moves by a stated amount — and calculates what that would mean for a defined set of holdings. Some tools also offer predefined stress cases, where the assumption is written in advance rather than chosen each time. Neither approach is more rigorous than the other; they answer different questions.
What a scenario result is and is not
A scenario answers one question: if this assumption held, what would it mean here? It does not say the assumption is likely or plausible. That judgement stays with whoever chose it, which is why a scenario result should never be quoted without its assumption beside it — a conditional figure detached from its condition reads as an expectation.
Mechanical scenarios
Some scenarios are arithmetic. Asset-price changes, exchange-rate changes and allocation or holding changes can be recalculated directly from the position. Results of this kind may be described as deterministic only when the holdings, prices, exchange rates, assumptions and timestamps are all defined: the same inputs then give the same answer every time, and the calculation can be checked by hand.
Model-based stress scenarios
Others are not arithmetic. Interest rates, volatility, macroeconomic conditions and combined stress cases do not have a single defined path into a portfolio's value; representing them requires modelling choices about how the change transmits. Describing these as deterministic alongside a price shock gives them an authority they have not earned, and hides the fact that a different reasonable choice would produce a different number.
Volatility is the clearest case
A volatility assumption may affect risk estimates, forecast distributions or model-based stress results, but it does not mechanically revalue a plain equity holding. A tool reporting a precise change in portfolio value from a volatility assumption alone is asserting more than the assumption supports, and the reader has no way to see which modelling choice produced it.
Using both
They complement each other because they start from opposite ends. A forecast describes what current conditions suggest; a scenario describes how much a particular change would matter to a given set of holdings. Learning that a portfolio moves little under an assumption you consider material is useful — and it is useful without requiring any forecast to be correct.
A fictional worked example
A fictional portfolio holds £8,000 in a US equity fund. A mechanical scenario applies a 10% fall in the fund's local-currency price, at stated prices and a stated exchange rate: the holding's contribution falls by £800, to £7,200. The arithmetic is reproducible and takes no view on whether such a fall is likely. Separately, a change in the sterling/dollar exchange rate would also affect the GBP result, and would be applied as its own stated assumption rather than folded into this one. All values are fictional and describe no real fund.
Common misunderstandings
Not quite: The scenario says my portfolio will fall by that amount.
It says that if the stated assumption held, at the stated prices and rates, the effect would be that. The assumption is an input, and the calculation takes no view on it.
Not quite: A scenario is a more cautious kind of forecast.
It is not a forecast. It carries no probability and makes no claim about likelihood; any caution has to come from how the assumption was chosen.
Not quite: All scenario results are exact.
Price, exchange-rate and allocation changes can be recalculated arithmetically. Rate, volatility and combined stress cases rest on modelling choices and should be read as estimates.
Limitations
- A scenario result is only as defined as its inputs: holdings, prices, exchange rates, assumptions and the time they were captured.
- Model-based stress results may omit second-order effects unless those effects are explicitly included in the published methodology.
- Nothing in a scenario estimates how likely its own assumption is.
- Where a tool offers predefined stress cases, each reflects whoever defined it, and that definition should be inspectable.
Sources and further reading
- Bank of England — stress testing the UK banking system (cited only to show that formal stress scenarios rest on explicit, published assumptions; Neo-Invest.AI does not follow this methodology and the citation is not an endorsement)
- Neo-Invest.AI Methodology — approved terminology for scenarios
Related explanations
Educational information only. This article does not constitute investment, tax or legal advice, and does not take your circumstances into account.