Value at Risk (VaR) explained: definition and trader use

By Ken Chigbo, Founder, KenMacro. Published 2026-05-13.

What is value at risk?

Value at Risk is a single number that summarises downside risk across a portfolio. It answers a precise question: over a given holding period, and at a given confidence level, what loss should not be exceeded under normal market conditions? A bank running a one day 95 percent VaR of two million pounds expects losses to breach that level on roughly five trading days out of a hundred. VaR is computed using three main approaches: historical simulation, the variance covariance method, and Monte Carlo simulation. Each makes different assumptions about return distributions, correlation stability, and tail behaviour, which produces materially different VaR estimates on the same book.

How traders use value at risk

Institutional desks use VaR as a daily risk budget. Trading limits at investment banks are typically expressed in VaR terms, with internal models reviewed by risk teams and validated against actual profit and loss through backtesting. Under the Basel framework, banks publish VaR backtest exceptions in their Pillar 3 disclosures, and persistent breaches force a higher capital multiplier. Retail traders rarely compute VaR explicitly, but the concept maps directly onto position sizing. A trader running fixed fractional risk of one percent per trade is implicitly setting a per trade VaR. Multi strategy retail portfolios can approximate VaR by simulating the worst rolling drawdown over recent history at a chosen percentile, which informs leverage caps and overnight exposure across correlated FX pairs.

Common misconceptions about Value at Risk

The first misconception is that VaR is a worst case loss. It is not. A 99 percent VaR says nothing about what happens on the worst one percent of days, where losses can be far larger. The second is that VaR is additive. It is not, because diversification means portfolio VaR is usually below the sum of component VaRs. The third is that VaR is model free. Every VaR figure depends on lookback window, weighting scheme, and distributional assumptions. The 2008 crisis exposed how variance covariance VaR understates tail risk when correlations break and volatility regimes shift, which is why Expected Shortfall is now preferred under Basel III FRTB.

Open a Vantage raw-spread account

Frequently asked

What is the difference between VaR and Expected Shortfall?

VaR gives the threshold loss at a confidence level, for example the loss that should be exceeded only one percent of the time. Expected Shortfall, also called Conditional VaR, gives the average loss in those tail scenarios beyond VaR. Expected Shortfall captures tail severity that VaR ignores, which is why the Basel Committee replaced VaR with Expected Shortfall at 97.5 percent confidence for market risk capital under the Fundamental Review of the Trading Book.

How is VaR calculated in practice?

Three methods dominate. Historical simulation revalues the current portfolio using actual returns from a lookback window, typically one to four years, then reads off the relevant percentile. The variance covariance method assumes returns are normally distributed and uses a covariance matrix to compute VaR analytically. Monte Carlo simulation generates thousands of synthetic return paths from a chosen model. Historical simulation is the most common because it avoids distributional assumptions and handles non linear payoffs naturally.

What time horizon should VaR use?

Horizon depends on use case. Trading desks typically compute one day VaR because positions can be unwound quickly. Asset managers often use ten day or one month horizons to match rebalancing cycles. Basel rules require ten day VaR for market risk capital, which is usually derived from one day VaR by scaling with the square root of time. That scaling assumes returns are independent and identically distributed, which breaks down during stressed regimes.

Is VaR useful for retail forex traders?

The full statistical apparatus is overkill for a single account, but the concept is directly relevant. A retail trader sizing positions at fixed percentage risk is setting a per trade VaR. Where retail traders benefit most is in thinking about portfolio VaR across correlated pairs. Holding long EUR/USD, long GBP/USD, and short USD/CHF simultaneously concentrates dollar risk in a way that single position sizing rules miss. A simple historical simulation across recent returns highlights that concentration.

Educational analysis only. Past performance does not guarantee future results. Manage risk against your own portfolio.