AIO.
EN ▾

Tools — Value at Risk (VaR) Calculator

Value at Risk (VaR) Calculator

Estimate the maximum expected loss on a portfolio at a chosen confidence level — using either the parametric (variance-covariance) method or your own historical return series.

Trade Setup

01
$
02
03
%
04
%
05

Results

Value at Risk ($) —
Value at Risk (%) —

Parametric VaR assumes normally distributed returns and scales volatility by √(horizon); Historical VaR reads the actual percentile from your pasted returns with no distribution assumption. Both express the expected loss magnitude, not a hard cap — losses can still exceed VaR in the tail beyond the chosen confidence level.

How to read Value at Risk

Value at Risk answers: "At this confidence level, how much could I lose over this time horizon?" A 95% 1-day VaR of $3,000 on a $100,000 portfolio means there's a 95% chance the portfolio won't lose more than $3,000 in a single day — equivalently, a 5% chance it does. Parametric VaR calculates this from volatility and a normal-distribution assumption, which is fast but can understate risk for assets with fat-tailed or skewed return distributions (common in crypto). Historical VaR instead reads the actual worst-case percentile directly from real past returns, capturing whatever distribution shape actually occurred, at the cost of needing a real return series and being limited by how much history you have. Neither method caps the loss — by definition, the confidence level's complement (5% for a 95% VaR) is exactly how often the true loss exceeds the VaR figure.

Frequently Asked Questions

What does a 95% VaR of $5,000 actually mean?
It means that, based on the model and data used, there is a 95% probability the portfolio's loss over the chosen horizon will not exceed $5,000 — and therefore a 5% probability that it will. VaR describes a threshold and a confidence level, not a worst-case ceiling.
Should I use Parametric or Historical VaR?
Parametric VaR is quick and only needs a volatility estimate, but it assumes returns follow a normal distribution — an assumption that tends to understate risk for assets with frequent large moves, like crypto. Historical VaR uses your own past returns with no distribution assumption, which captures real fat tails and skew, but its accuracy depends on having enough representative historical data.
Why does the VaR percentage scale with the square root of time?
Under the common assumption that returns are independent from period to period, volatility (standard deviation) grows with the square root of time while the mean return grows linearly with time — this is why the volatility term in parametric VaR is multiplied by √(horizon) rather than by the horizon directly.

Related trading tools

Trading Expectancy & Kelly Calculator Compound Growth Calculator Drawdown Recovery Calculator DCA Calculator Average Down Calculator Risk of Ruin Calculator
Start 5-Day Free Trial