Value at Risk (VaR) Calculator — User Guide & Reference
Value at Risk condenses a position's danger into one sentence: there is a 95% chance this position loses no more than some amount over the next N days. This guide explains the formula, every input, and a worked example.
Disclaimer: This guide is provided for education only and does not constitute investment advice.
What VaR estimates
VaR is a loss threshold at a confidence level over a horizon. A 95% one-day VaR of $1,000 means losses should stay within $1,000 on 95% of days under the model.
The formula
VaR = position times volatility times z-score times square root of days. The z-score depends on the confidence level (1.645 for 95%). Time scaling uses the square root of days.
Worked example: a $10,000 position
You hold a $10,000 position with 2% daily volatility and want a one-day, 95% VaR. The 95% one-tailed z-score is 1.645, so one-day VaR = 10,000 times 2% times 1.645, or $329. For five days, multiply by the square root of five: about $736.
Why the normal assumption is the weakness
Markets have fat tails: real losses exceed the normal-model VaR more often than the model predicts. That is why the calculator pairs VaR with CVaR — the expected loss in the tail beyond the threshold.
How to use the calculator
Enter the position amount, volatility, horizon, and confidence level. Read VaR and CVaR together. Use VaR for comparison and ranking, not as a claim of the most you can lose.
Common mistakes
The most common mistake is trusting VaR as an absolute upper bound on losses. Another is assuming returns are normally distributed. And scaling daily VaR linearly over time is wrong.
Parameters explained
| Parameter | What it means |
|---|---|
| Position amount | The size of the position at risk. |
| Volatility / horizon | Expected volatility and the number of days. |
| Confidence level | 90%, 95% or 99% — the probability threshold. |
Key concepts
Confidence level
The probability threshold — 95% VaR means losses should exceed VaR only 5% of the time under the model.
Volatility input
The standard deviation of returns drives the estimate.
Time scaling
Loss scales with the square root of time under the normal model.
Mistakes traders make
✗ Trusting VaR as an absolute upper bound on losses
✓VaR is a threshold under a model, not a guarantee. The 5% tail beyond it can be severe — exactly what CVaR tries to capture.
✗ Assuming returns are normally distributed
✓Markets have fat tails. Real losses exceed the normal-model VaR more often than the model predicts.
✗ Scaling daily VaR linearly over time
✓Under the standard model risk scales with the square root of time, not linearly — but that too is a modeling convenience, not a market fact.
When this tool fits — and when it does not
Use it when
- ✓Summarizing a single position's risk into one comparable number.
- ✓Comparing the riskiness of different positions or assets.
- ✓Setting a daily loss threshold as part of a risk discipline.
Skip it when
- ✗Relying on it for tail-risk management without CVaR.
- ✗Estimating portfolio risk without accounting for correlations.
- ✗Trading a position whose VaR you have not bothered to measure.
Practical tips that actually help
Always pair VaR with CVaR
VaR tells you the boundary; CVaR tells you how bad it gets beyond the boundary. The second number is the more valuable one.
Use it for comparison, not prediction
Ranking positions by VaR is defensible; quoting it as 'the most I can lose' is not.
Re-estimate when volatility changes
VaR is only as fresh as its volatility input. After a volatility expansion, the old number understates real risk.
Related & next steps
Frequently asked questions
What is VaR's main limitation?▾
It assumes a distribution (often normal) and says little about extreme tail losses. CVaR helps by focusing on losses beyond VaR.
What does 95% VaR actually promise?▾
Under the model's assumptions, losses should stay within the VaR figure 95% of the time. It does not promise the other 5% is mild.
Why does VaR scale with the square root of time?▾
Under the standard model, random daily moves partly cancel out over longer horizons, so risk grows with the square root of days.
Is VaR meant for a single position or a whole portfolio?▾
It works for both, but the calculator measures one position. A portfolio version would also need correlations.
How is VaR different from max drawdown?▾
Max drawdown is a historical extreme; VaR is a forward-looking estimate of a loss threshold under a distribution.
Why is CVaR more useful than VaR?▾
VaR tells you the boundary; CVaR tells you how bad it gets beyond the boundary. The second number is the more valuable one.
- Type
- User guide
- Updated
- 2026-09
- Source
- SoloTrade.io educational content
VaR is a statistical model with assumptions and does not guarantee actual losses stay within the estimate.