Value at Risk Calculator

Enter portfolio value, volatility and confidence to compute Value at Risk.

How it is calculated

The tool computes parametric (variance-covariance) Value at Risk: VaR = V × z × σ × √T, and the same as a share of the portfolio, VaR% = z × σ × √T × 100. V is the portfolio value, σ is the volatility per period (you enter it as a percentage, so 1.2 means 1.2%), T is the holding period counted in the same periods as σ, and z is the one-tailed normal value for the confidence level: 1.2816 for 90%, 1.6449 for 95%, 2.3263 for 99%. Any other confidence value falls back to 95%. Value and volatility cannot be negative and the horizon must be above zero. It assumes normally distributed returns and ignores extreme events and the expected return.

Worked example

A portfolio worth $100,000 has a daily volatility of 1.2%. You want the 10-day loss that should not be exceeded with 99% confidence. Step 1: σ = 0.012 and √10 = 3.1623. Step 2: z for 99% = 2.3263. Step 3: VaR = 100,000 × 2.3263 × 0.012 × 3.1623 = 8,827.87. The tool shows $8,827.87, which is 8.83% of the portfolio. At 95% confidence, with z = 1.6449, the same inputs give $6,241.78.

Try this example →

Common mistakes

  • Mixing periods: entering an annual volatility with a horizon in days, or the reverse, when σ and T must be in the same unit (daily σ with days, annual σ with years).
  • Typing volatility as 0.012 instead of 1.2; the field is a percentage, so 0.012 means 0.012% and VaR nearly vanishes.
  • Reading VaR as the worst possible loss: at 99% confidence, a loss bigger than the figure is still expected on about 1 period in 100.
  • Choosing a confidence value that is not in the list (for example 97.5), which the tool quietly replaces with 95%.
  • Using the result for portfolios with options or heavy tails, where the normal assumption underestimates risk.

About this calculator

This calculator uses the parametric (variance-covariance) VaR method: VaR = portfolio value × z-score for the chosen confidence level × volatility × √holding period. The z-score comes from the standard normal distribution (1.645 for 95% confidence, 2.326 for 99%), and the √time scaling assumes returns follow a random walk, so risk grows with the square root of the holding period rather than linearly.

Risk managers and portfolio managers use this to state, in dollar terms, the maximum loss a portfolio is unlikely to exceed over a given horizon at a given confidence level — a figure that feeds directly into regulatory capital calculations under frameworks like Basel — while traders use it day to day to size positions and keep risk within a desk's mandate.

Was this helpful?

You can pick up to 3 reactions.

Comments

  • Be the first to comment.

Popular calculators

All Calculators