Glossary term
Historical Simulation
Historical simulation is a risk-modeling method that estimates possible portfolio outcomes by replaying actual past market changes through today's portfolio.
Updated
Read time
What Is Historical Simulation?
Historical simulation is a risk-modeling method that estimates possible portfolio outcomes by replaying actual past market changes through today's portfolio. Instead of assuming returns follow a normal distribution, the method asks what would happen if the portfolio experienced a set of historical shocks again.
Historical simulation is often used in value-at-risk, stress testing, scenario analysis, and risk-management reporting. Its appeal is intuitive: it uses real market moves. Its weakness is also intuitive: the future may not look like the sample period.
Key Takeaways
- Historical simulation applies past market changes to a current portfolio.
- It can capture actual historical co-movements across assets without assuming a simple normal distribution.
- The result depends heavily on the chosen lookback window and data quality.
- It can miss risks that have not appeared in the historical sample.
How Historical Simulation Works
A risk team gathers a history of market moves, such as daily changes in interest rates, equity prices, credit spreads, currencies, or commodity prices. It then applies those historical moves to today's positions. Each day in the historical window becomes a possible scenario for the current portfolio.
The resulting distribution can be used to estimate losses at different confidence levels. For example, if the model uses 1,000 historical daily scenarios, the 1% worst outcome is read from the bottom tail of those simulated results. The method is simple to explain because each scenario comes from a real historical observation.
Where It Helps
Historical simulation can capture relationships that are hard to model cleanly. During a past crisis, stocks, credit spreads, currencies, and rates may have moved together in ways that a simple model would miss. Replaying that period can show how a current portfolio might behave if similar conditions returned.
It is also useful for communicating risk. A board or investment committee may understand a historical stress scenario more easily than a purely statistical model.
Key Modeling Choices
Choice | Why it matters |
|---|---|
Lookback window | A short window may miss crises; a long window may include stale regimes. |
Frequency | Daily, weekly, or monthly data can produce different risk estimates. |
Weighting | Equal weighting treats old and recent observations the same; decay methods do not. |
Portfolio mapping | Complex positions need accurate sensitivity or full revaluation. |
Where It Can Mislead
Historical simulation can be too comforting when the sample period was calm. It can also be too pessimistic if an old crisis no longer matches the portfolio, market structure, or policy environment. New instruments, crowded trades, leverage, and liquidity stress may behave differently from the past.
The method also does not automatically explain cause. It shows what would happen if a set of market moves repeated, but it does not say whether those moves are likely or what might trigger them.
The Bottom Line
Historical simulation is a practical way to estimate portfolio risk by replaying real past market shocks through current holdings. It is useful because it is grounded in actual market behavior, but it should be paired with judgment, stress tests, and awareness of risks that history has not yet shown.