Historical var

• [TERM] = Historical VaR
• [RELATED_TERM] = Parametric VaR
• [TERM_CATEGORY] = Risk Management

</hidden table>

What Is Historical VaR?

Historical VaR, or Historical Value at Risk, is a non-parametric method used in Risk Management to estimate potential financial losses over a specific period. It directly uses past market data to predict future potential losses by re-ordering historical returns from worst to best. This approach falls under the broader category of Value at Risk (VaR) methodologies, providing a statistical measure of the maximum expected loss within a given confidence level and time horizon, based on actual past performance. Historical VaR is particularly useful for financial institutions and portfolio managers seeking to understand their risk exposure without making assumptions about the underlying statistical distribution of returns.

History and Origin

The concept of Value at Risk gained significant traction in the early 1990s, particularly after the release of J.P. Morgan's RiskMetrics system in 1994. J.P. Morgan developed an internal firm-wide VaR system in the late 1980s.¹⁸ This initiative, spearheaded by Chairman Dennis Weatherstone's request for daily risk reports, led to the eventual public release of the risk metrics methodology. The goal was to promote greater transparency in market risk and establish a benchmark for measurement.¹⁷ The RiskMetrics methodology itself included the historical simulation approach, which became an industry standard for calculating VaR.¹⁶ This method allowed market participants to utilize historical data sets for risk assessment, contributing to the widespread adoption of VaR as a key risk measurement tool.¹⁵

Key Takeaways

• Historical VaR uses actual past returns to estimate potential future losses.
• It is a non-parametric method, meaning it does not assume a specific distribution for returns.
• Historical VaR provides a single number representing the maximum expected loss at a given confidence level over a specific period.
• It is widely applied in portfolio management and regulatory capital calculations for financial institutions.
• The method can be sensitive to the length and representativeness of the historical data period chosen.

Formula and Calculation

The calculation of Historical VaR does not rely on a complex mathematical formula in the traditional sense, but rather a simple, direct methodology based on historical data.

The steps are as follows:

1. Collect Historical Data: Gather a time series of historical returns for the asset or portfolio in question. This typically involves daily returns over a chosen lookback period (e.g., 250 days, 500 days).
2. Order Returns: Sort the collected historical returns from the smallest (largest negative) to the largest (largest positive).
3. Determine Confidence Level: Choose a desired confidence level (e.g., 95%, 99%). This indicates the probability that the actual loss will not exceed the calculated VaR.
4. Identify the VaR Value:
   • For a given confidence level (\alpha), the Historical VaR is the return at the ((1-\alpha)) percentile of the sorted historical returns.
   • For example, for a 95% confidence level, you would find the 5th percentile of the sorted returns. For a 99% confidence level, you would find the 1st percentile.

The Historical VaR is given by:

Where:

• (\alpha) = confidence level (e.g., 0.95 for 95%)
• Percentile() refers to the value at that specific percentile in the sorted distribution of returns.

This process essentially looks at the worst-case scenarios that occurred historically within the chosen confidence interval. The historical returns often refer to the daily percentage change in the value of an asset or portfolio.

Interpreting the Historical VaR

Interpreting Historical VaR involves understanding what the calculated number signifies in practical terms for financial institutions and investors. A Historical VaR of, for example, $1 million at a 99% confidence level over a one-day horizon means that, based on historical data, there is a 1% chance that the portfolio could lose $1 million or more within a single day. Conversely, it implies that 99% of the time, the portfolio's daily losses are expected to be less than $1 million.

This metric provides a quantifiable measure of downside market volatility. It helps risk managers assess the potential impact of adverse market movements on their trading portfolios. However, it's crucial to remember that Historical VaR is backward-looking; it assumes that past performance is indicative of future risk, which may not always hold true, especially during periods of unprecedented market conditions.

Hypothetical Example

Consider a hypothetical investment portfolio with a current value of $10 million. To calculate the 1-day Historical VaR at a 95% confidence level, an analyst collects the daily percentage returns for the past 250 trading days.

1. Collect Data: The analyst has 250 daily percentage returns for the portfolio.
2. Sort Returns: The returns are sorted in ascending order (from most negative to most positive).
   • Example returns (simplified for illustration, in %): -3.5%, -2.8%, -2.1%, -1.9%, -1.5%, ..., 0.5%, 1.2%, 2.0%
3. Determine Position: For a 95% confidence level, the percentile needed is the 5th percentile (100% - 95%). With 250 observations, the position corresponding to the 5th percentile is (250 \times 0.05 = 12.5). Since it's between two values, we'd typically take the 13th worst return in the sorted list (rounding up).
4. Identify VaR: Let's assume the 13th worst return in the sorted list is -1.8%.

Therefore, the 1-day Historical VaR at a 95% confidence level for this portfolio is 1.8% of the portfolio's value. This translates to a potential loss of (0.018 \times $10,000,000 = $180,000). Based on the last 250 trading days, there is a 5% chance that the portfolio could lose $180,000 or more within a single day. This illustrates how the method provides a clear, digestible number for risk assessment.

Practical Applications

Historical VaR is a widely used tool in various areas of finance for assessing and managing risk. Its primary applications include:

• Regulatory Capital Calculation: Many regulators, including those under the Basel Accords, have historically allowed banks to use their internal VaR models, including historical simulation, to calculate minimum capital requirements for market risk.¹³, ¹⁴ This ensures that financial institutions hold sufficient capital to cover potential trading losses.¹² The Basel Committee on Banking Supervision's framework, including Basel III reforms, emphasizes strong risk management and capital adequacy.
• Risk Reporting: Financial firms use Historical VaR to report daily risk exposures to senior management, boards of directors, and regulators. This provides a clear, consolidated view of potential losses across diverse trading portfolios.
• Portfolio Risk Management: Portfolio managers utilize Historical VaR to understand the downside risk of their investments. It helps them set appropriate risk limits, optimize asset allocation, and evaluate the risk-return trade-off of different strategies.
• Investment Due Diligence: Investors and analysts may use Historical VaR as part of their due diligence process to evaluate the risk characteristics of a fund or investment vehicle before committing capital.
• Limit Setting: Banks and other financial entities establish trading limits based on Historical VaR to control the overall risk taken by individual traders or desks.

Limitations and Criticisms

While Historical VaR is a popular and intuitive risk measurement technique, it has several notable limitations and has faced criticism, especially in the wake of financial crises.

• Reliance on Historical Data: The core assumption of Historical VaR is that future market movements will resemble past ones. This can be a significant drawback during periods of extreme market stress or structural changes, as historical data may not capture unprecedented events.¹¹ The 2008 financial crisis highlighted this weakness, as many VaR models failed to account for the interconnectedness and systemic risks that emerged.⁸, ⁹, ¹⁰
• Inability to Capture "Black Swan" Events: Historical VaR struggles to account for rare, high-impact events, often termed "black swan" events, that have no precedent in the historical data used for its calculation.⁶, ⁷ It focuses on events within a specific confidence interval but provides limited information about losses beyond that threshold, known as tail risk.⁴, ⁵
• Limited Forward-Looking Insight: Unlike some other risk models, Historical VaR does not incorporate expectations of future market conditions or changes in economic fundamentals. It is purely backward-looking.
• Data Intensive: Accurate Historical VaR calculations require a significant amount of reliable historical data, which may not always be available for new or thinly traded assets.
• "False Sense of Security": Critics argue that an over-reliance on a single VaR number can create a false sense of security for risk managers, leading to complacency and insufficient preparedness for extreme market events.³

Regulatory bodies, such as the Federal Reserve, have acknowledged these limitations and emphasized the need for complementary stress testing and other risk measures to provide a more comprehensive view of risk.¹, ²

Historical VaR vs. Parametric VaR

Historical VaR and Parametric VaR are two distinct approaches to calculating Value at Risk, primarily differing in their underlying assumptions about data distribution.

While Historical VaR is straightforward due to its reliance on observed data, Parametric VaR can be more flexible, especially when historical data is scarce or when specific distributional properties are known. However, the accuracy of Parametric VaR heavily depends on the validity of its distributional assumptions, which often come under scrutiny during periods of market turmoil.

FAQs

What is the main advantage of Historical VaR?

The main advantage of Historical VaR is its simplicity and its non-parametric nature. It does not require any assumptions about the statistical distribution of returns, as it directly uses observed past data. This can be beneficial when dealing with assets or portfolios whose returns do not follow a standard distribution.

How is the confidence level used in Historical VaR?

The confidence level in Historical VaR defines the probability that the actual loss will not exceed the calculated VaR. For example, a 99% confidence level means that, based on historical data, there is a 1% chance of a loss equal to or greater than the Historical VaR. It helps pinpoint the specific percentile of worst-case scenarios from the sorted historical returns.

What is a typical lookback period for Historical VaR?

Common lookback periods for Historical VaR range from 250 days (approximately one trading year) to 500 days or more. The choice of lookback period is crucial as it directly impacts the VaR estimate. A longer period provides more data but might include less relevant past conditions, while a shorter period might be more responsive to recent market conditions but prone to noise.

Does Historical VaR predict future losses with certainty?

No, Historical VaR does not predict future losses with certainty. It provides a statistical estimate of potential losses based on past data and a given confidence level. It's a risk management tool that helps quantify risk exposure but cannot account for all unforeseen market events or guarantee specific outcomes.