What confidence level should I use for VaR?

Depends on risk tolerance and purpose: 90% suitable for aggressive investors seeking high returns, 95% suitable for average investors balancing risk and return, 99% suitable for institutional investors requiring strict risk control.

What is the difference between VaR and Maximum Drawdown?

VaR focuses on the maximum daily loss at a specific confidence level (short-term 1-10 days), while Max Drawdown focuses on the maximum decline over the entire investment period (peak to trough). They measure different timeframes and focus points.

VaR may fail during black swan events (unpredictable extreme conditions), sudden market structure changes (e.g., 2020 circuit breakers), liquidity crises (unable to sell at market prices), and model risk (assumptions don't match reality). It is recommended to use Stress Tests and CVaR alongside VaR.

Which VaR calculation method should I use?

Beginners can use historical simulation for its simplicity, most users can use the parametric method for quick estimation, and advanced investors with complex portfolios can use Monte Carlo for greater accuracy. The three methods can be cross-validated.

Can VaR be used for options?

Yes, but be aware of non-linear characteristics. Linear products like stocks and bonds can use the parametric method, while non-linear products like options and structured products are better suited to Monte Carlo, because the relationship between option prices and underlying assets is non-linear.

Compare predicted VaR with actual losses. Common methods include the Kupiec POF Test (checks VaR breach frequency) and the Traffic Light Test (classifies breaches into green/yellow/red). If actual breaches exceed expectations, the model is underestimating risk.

VaR (Value at Risk) Calculation Guide 2026: A Must-Learn Risk Quantification Tool for Retail Investors

VaR Core Definition

VaR (Value at Risk) is a risk quantification metric widely used in the financial industry, answering one core question: "Under normal market conditions, what is the maximum loss my portfolio could suffer?"

Three Elements of VaR

Element	Description	Common Setting
Time Horizon	Time period for risk assessment	1 day, 10 days, 1 month
Confidence Level	Statistical certainty	95%, 99%
Amount/Percentage	Maximum possible loss	Absolute amount or percentage

Example: 95% confidence level, 1-day VaR = $50,000 → There is a 95% probability that tomorrow's loss will not exceed $50,000 → In other words, there is a 5% probability that the loss will exceed $50,000

Three VaR Calculation Methods

Method 1: Historical Method (Simplest)

Directly use historical data to calculate:

Collect daily returns from the past N days
Sort returns from smallest to largest
Find the 5th percentile (95% confidence) or 1st percentile (99% confidence)

Example: Among the past 100 days' returns, the 5th worst return is -3% → 95% confidence, 1-day VaR = 3%

Pros: Simple and intuitive, no distribution assumptions needed Cons: Assumes history will repeat itself

Method 2: Variance-Covariance Method

Assumes returns follow a normal distribution:

VaR = Investment Amount × Z-value × Volatility

Z-value: 95% confidence = 1.65, 99% = 2.33
Volatility: Historical volatility (standard deviation)

Example: Invest $1M, volatility 2%, 95% confidence → VaR = $1M × 1.65 × 2% = $33,000

Pros: Fast calculation, suitable for portfolios Cons: Assumes normal distribution (actual markets have fat tails)

Method 3: Monte Carlo Simulation

Use computers to randomly simulate a large number of possible scenarios:

Set return distribution parameters
Randomly simulate 10,000+ price paths
Analyze loss distribution to find VaR

Pros: Most accurate, can handle complex scenarios Cons: Requires computing power, difficult for retail investors

How Retail Investors Can Apply VaR

Application 1: Assess Single Stock Risk

Calculate VaR for each holding to understand individual stock risk exposure:

Stock A: 1-day VaR = 5%
Stock B: 1-day VaR = 3% → Stock A has higher risk, should reduce position

Application 2: Total Portfolio Risk

Calculate the VaR of the entire portfolio, considering correlations between stocks:

Portfolio VaR < Sum of individual stock VaRs (due to diversification)
If portfolio VaR is close to the sum of individual VaRs, diversification is poor

Application 3: Stress Testing

Under extreme market conditions (e.g., 2008 Financial Crisis, 2020 Pandemic), VaR may fail. In such cases, stress testing is needed:

Assume market drops of 20%, 30%, 40%
Calculate portfolio losses under these scenarios
Assess whether you can withstand them

For more stress testing techniques, see Portfolio Stress Testing.

Limitations of VaR

Limitation 1: Does Not Predict Extreme Events

VaR assumes "normal market conditions" and cannot predict black swan events. In 2008, many banks' 99% VaR models failed.

Limitation 2: Doesn't Tell You How Much You'll Lose (When Exceeding VaR)

VaR only says "95% probability loss won't exceed X," but doesn't say how much you'll lose in that 5% case. For this, you need Conditional VaR (CVaR).

Limitation 3: Assumes History Repeats

All VaR calculations are based on historical data. If market structure changes, historical VaR may not apply. It is recommended to use Regime & Risk analysis to dynamically adjust risk assumptions.

Summary

Core value of VaR:

Quantify risk — Use numbers, not feelings
Compare across portfolios — Can compare risk across different portfolios
Set limits — As a reference indicator for risk control

But remember: VaR is a tool, not a crystal ball. Use it with diversification and position sizing to build a complete risk management system. For more quantitative risk management knowledge, visit the Learning Center.