PortfoliosLab logoPortfoliosLab logo
...

Risk Analysis

/

Value at Risk

Value at Risk

Learn how to estimate potential one-day downside under normal conditions with the Value at Risk (VaR) tool.

Risk Metrics
Downside Risk
Last updated: February 21, 2026

Value at Risk (VaR) estimates a loss threshold for a selected confidence level over the next trading day.

In simple terms, VaR answers: "How much could this portfolio lose in a bad day with a chosen probability?"

This tool helps you:

  • quantify short-horizon downside in dollar terms
  • compare risk under different methods and confidence levels
  • monitor how potential losses change through time

VaR converts abstract volatility into a concrete potential-loss estimate, which is useful for risk budgeting, position sizing, and stress awareness.


How to Use the Tool

Use this workflow in Value at Risk:

1

Select Portfolio Positions

Build or choose the portfolio in the portfolio selector before calculation.

2

Choose VaR Method

Select Historical VaR or Gaussian VaR depending on how you want tail risk modeled.

3

Set Significance Level

Enter the significance level (for example, 5% or 1%). Lower levels represent more extreme downside thresholds.

4

Set Current Portfolio Value

Provide current portfolio value so VaR is calculated in absolute dollar terms.

5

Calculate and Interpret the Rolling Chart

Click "Calculate VaR" and review how one-day potential loss evolves over time.

Value at Risk settings with method, significance level, current portfolio value, and calculate button
Practical Tip

Run the same portfolio with multiple significance levels (for example, 5% and 1%) to see how sensitive downside estimates are to tail assumptions.


Tool Settings

The VaR tool has three core settings:

Method

Historical VaR uses empirical return history; Gaussian VaR assumes a normal return distribution.

Significance Level

Defines tail probability. A 5% level means a 5% chance of losses worse than the VaR threshold over one day.

Current Portfolio Value

Converts relative risk estimate into an absolute dollar loss threshold.

Method guidance:

  • Historical VaR: more data-driven and sensitive to observed tail behavior.
  • Gaussian VaR: smoother and simpler, but may underestimate risk in fat-tail regimes.

If required inputs are missing (for example, invalid positions), calculation is blocked until validation errors are resolved.


Results: Section-by-Section Guide

Rolling VaR Chart

This is the main output section. It shows how estimated one-day VaR changes over time.

Use it to identify:

  • periods of rising or falling downside pressure
  • whether risk is stable or regime-dependent
  • how method/level choices affect the magnitude of estimated loss
Rolling Value at Risk chart over time
Interpretation Framework

VaR is a threshold estimate, not a worst-case bound. It tells you how bad losses could be up to a chosen probability level, not what happens beyond that threshold.


Example

Suppose you calculate VaR for a portfolio at a 5% significance level and get $24,592.

Interpretation:

  • There is a 5% chance that next-day loss exceeds $24,592.
  • In 95% of days, loss is expected to be smaller than that threshold (under model assumptions).

Important limitation:

VaR does not describe how severe losses are once the threshold is breached. For tail severity analysis, use Expected Shortfall.


Best Practices

Compare Historical and Gaussian results

Differences between methods often reveal distribution-shape risk.

Use multiple significance levels

Check both moderate and extreme-tail views (for example, 5% and 1%).

Monitor rolling behavior, not one value

Risk profile changes with market regime; re-evaluate VaR over time.

Pair VaR with Expected Shortfall

Use ES to evaluate average severity when losses exceed VaR.

Was this article helpful?