What is Value at Risk (VaR)?

Value at Risk is a statistical measure that quantifies the maximum expected loss over a specific time horizon at a chosen confidence level, such as 95% or 99%.

What is the formula for calculating Value at Risk?

Under the parametric normal method, VaR = - (mu - z * sigma) * Portfolio Value, where mu is the expected return, sigma is standard deviation, and z is the critical value for the chosen confidence level (1.645 for 95%, 2.326 for 99%). Under historical simulation, VaR is the negative value of the (1 - confidence level) percentile of historical returns.

What is the difference between 95% VaR and 99% VaR?

A 95% VaR estimates the loss threshold that will be exceeded 5% of the time (roughly 1 in 20 periods), while a 99% VaR represents a stricter cutoff exceeded only 1% of the time (roughly 1 in 100 periods). The 99% VaR is always higher in magnitude than the 95% VaR.

How do traders use VaR in strategy backtesting?

Traders use backtest VaR to determine realistic capital allocation, set maximum position sizes, verify that outlier losses fit within risk tolerances, and establish margin safety buffers before deploying live capital.

What are the main drawbacks of Value at Risk?

VaR does not describe tail severity beyond the cutoff, can underestimate extreme fat-tail risk during market crises, and can be sensitive to short sample periods.

Backtesting guide

Value at Risk (VaR) in Trading & Backtesting

Value at Risk (VaR) is a quantitative risk metric that estimates the maximum potential loss a portfolio or trading strategy is expected to incur over a specified time horizon at a given confidence level under normal market conditions.

Read the practical guide

Key Takeaways

VaR specifies three core variables: the potential loss amount, the time horizon, and the confidence level (such as 95% or 99%).
The metric can be calculated through historical simulation, parametric variance-covariance formulas, or Monte Carlo modeling.
A 95% daily VaR of $1,000 means that on 95 out of 100 trading days, losses should not exceed $1,000.
VaR does not describe the magnitude of loss once the threshold is breached, which is why Expected Shortfall (CVaR) is used alongside it.

What is Value at Risk and how it works

Value at Risk provides a single summary statistic of downside exposure across a defined holding period. Financial institutions, prop firms, and systematic traders use VaR to monitor leverage, allocate capital, and establish risk budgets. Rather than looking only at worst-case single events, VaR quantifies regular statistical loss boundaries based on empirical or modeled return distributions.

Time Horizon: The holding period over which risk is measured (such as 1 day, 10 days, or 1 month).
Confidence Level: The statistical probability that losses remain within the calculated threshold (commonly 95% or 99%).
Loss Metric: Expressed as a percentage of capital or as an absolute dollar amount.

Value at Risk formulas and calculation methods

Traders and risk managers calculate VaR using three primary methodologies. The parametric method assumes normally distributed returns and applies standard deviation multipliers (z-scores). The historical method sorts observed past returns directly without assuming a theoretical distribution curve. Monte Carlo simulation generates thousands of random price paths from stochastic parameters.

Parametric VaR Formula: VaR = - (Mean Return - Z-Score * Standard Deviation) * Portfolio Value.
Historical VaR Method: Rank historical percentage returns from worst to best and identify the percentile cutoff (for example, the 5th percentile for 95% confidence).
Monte Carlo Simulation: Simulate price paths with assumed volatility and correlation structures, then take the percentile loss across generated paths.

How to calculate Value at Risk from backtest trade logs

When evaluating algorithmic trading strategies, historical trade-by-trade returns provide the empirical sample needed to measure strategy VaR. A strategy with a strong win rate can still carry an unacceptable 99% VaR if occasional outlier losses wipe out weeks of steady gains. Reviewing trade distributions reveals whether sample size is large enough to produce reliable percentile estimates.

Export closed trade returns and trade durations from your backtesting platform.
Sort trade returns in ascending order to extract empirical percentile loss thresholds.
Compare 95% and 99% VaR levels against account stop-out rules and margin limits.
Cross-reference VaR with maximum peak-to-trough drawdown to evaluate multi-trade recovery cycles.

Limitations of VaR and when to pair it with CVaR

While VaR is an industry standard, it has well-documented weaknesses. It assumes stable distribution characteristics that can break down during liquidity shocks and market crashes. Most importantly, VaR provides no information about the size of losses in the remaining tail. For example, a 95% VaR tells you what happens on 95% of days, but tells you nothing about the severity of losses on the worst 5% of days.

Fat-tail events occur more frequently in financial markets than normal distribution models assume.
VaR is not a sub-additive risk measure, meaning portfolio VaR can occasionally exceed the sum of individual component VaRs.
Pair VaR with Conditional Value at Risk (CVaR) to measure the average loss in the extreme tail.

Direct Pineify workflow

Calculate VaR and loss distributions from your backtest

Pineify Backtest Deep Report processes your TradingView trade CSV export to calculate 95% and 99% Value at Risk, drawdown durations, and return distribution metrics automatically.

Backtest Deep Report

Upload your TradingView CSV export to audit trade VaR, distribution skewness, and equity curve stability.

Analyze backtest VaR

Boundary: Backtest Deep Report evaluates historical trade records; it does not predict future market volatility or eliminate trading risk.

Primary sources

Philippe Jorion, Value at Risk: The New Benchmark for Managing Financial RiskVerified 2026-09-02
SEC Investor.gov, Performance ClaimsVerified 2026-08-16

Related Resources

Conditional Value at Risk (CVaR)Tail Risk in Trading Risk-Adjusted Return Metrics Backtest Statistical Significance Backtest Deep Report

This page is educational and does not provide investment advice. Backtests are hypothetical, depend on their data and assumptions, and do not guarantee future results. Trading can result in substantial loss.