Monte Carlo Simulation
Monte Carlo simulation generates thousands of possible future scenarios by randomly sampling from historical return distributions. This helps you understand the range of possible outcomes, not just the expected outcome.
How It Works
1. Select instruments and allocation weights
2. Choose the number of simulations (default: 1,000) and time horizon
3. The system randomly samples daily returns from historical data and projects them forward
4. Results show the distribution of possible portfolio values at the end of the horizon
Output Metrics
| Metric | Description | How to Interpret |
|---|---|---|
| Median Outcome | The 50th percentile result - half of simulations did better, half did worse | This is your "most likely" outcome, more reliable than the mean which can be skewed by outliers |
| 5th Percentile (VaR) | The worst 5% of outcomes - your downside risk estimate | If the 5th percentile shows -30%, there is roughly a 5% chance of losing 30% or more |
| 95th Percentile | The best 5% of outcomes - your upside potential | Useful for setting optimistic but not unrealistic expectations |
| Mean Return | Average across all simulations | Can be skewed by extreme outliers; prefer the median |
| Standard Deviation | Spread of outcomes | Wider spread = more uncertainty about the final result |
| Probability of Loss | Percentage of simulations that ended with a loss | Below 30% is generally acceptable for growth portfolios; above 50% is a warning |
Fan Chart
The fan chart shows the distribution of outcomes over time:
- Dark band (25th-75th percentile): The "likely" range
- Light band (5th-95th percentile): The "possible" range
- Median line: The central tendency
Limitations
Monte Carlo assumes that future returns are drawn from the same distribution as historical returns. It does not account for regime changes, black swan events, or structural shifts. Use it as one input among many, not as a prediction.