Understanding Sigma Events
Sigma events represent extreme price deviations. They measure how far a price moves from its mean, expressed in standard deviations. A 1-sigma event means the price moved one standard deviation from the mean. A 2-sigma event means two standard deviations. Higher sigma values indicate rarer, more significant deviations. Traders use sigma events to identify potential mean reversion opportunities.
Statisticians assume asset prices follow a normal distribution. For a normal distribution:
- 68.2% of data falls within +/- 1 standard deviation (1-sigma).
- 95.4% of data falls within +/- 2 standard deviations (2-sigma).
- 99.7% of data falls within +/- 3 standard deviations (3-sigma).
This implies that price movements exceeding 2 or 3 standard deviations occur infrequently. These infrequent events often precede a return to the mean.
Calculating Sigma Deviations
Calculating sigma deviations requires a mean and a standard deviation. Traders typically use a moving average for the mean and a moving standard deviation for the volatility measure.
Consider Apple stock (AAPL) on October 26, 2023. Assume a 20-day simple moving average (SMA) for AAPL is $170.00. Assume a 20-day standard deviation for AAPL is $5.00.
On October 26, 2023, AAPL closes at $165.00. Deviation from mean = $165.00 - $170.00 = -$5.00. Sigma deviation = Deviation / Standard Deviation = -$5.00 / $5.00 = -1.0 sigma.
This represents a 1-sigma event below the mean. It suggests a moderate deviation.
Now, consider a hypothetical scenario for AAPL on October 27, 2023. SMA remains $170.00. Standard Deviation remains $5.00. AAPL closes at $155.00. Deviation from mean = $155.00 - $170.00 = -$15.00. Sigma deviation = -$15.00 / $5.00 = -3.0 sigma.
This -3.0 sigma event indicates a significant, rare deviation. Many mean reversion strategies target such extreme events.
The choice of lookback period for the moving average and standard deviation impacts the sigma calculation. Shorter periods react faster to recent price changes. Longer periods provide a smoother, less reactive measure. Traders must backtest different lookback periods to optimize their strategy.
Trading Strategies Using Sigma Events
Mean reversion strategies often trigger trades based on sigma thresholds. A common strategy involves buying assets that fall below a certain negative sigma threshold and selling assets that rise above a positive sigma threshold.
Example: A 2-sigma mean reversion strategy. Entry condition: Buy when price falls below -2 standard deviations from its moving average. Exit condition: Sell when price returns to the moving average (0 sigma) or crosses above it.
Consider the SPDR S&P 500 ETF (SPY). Assume a 50-day SMA for SPY is $440.00. Assume a 50-day standard deviation for SPY is $8.00.
Lower 2-sigma band = $440.00 - (2 * $8.00) = $440.00 - $16.00 = $424.00. Upper 2-sigma band = $440.00 + (2 * $8.00) = $440.00 + $16.00 = $456.00.
On March 13, 2023, SPY closed at $385.67. The 50-day SMA was approximately $400.00. The 50-day standard deviation was approximately $7.00. Lower 2-sigma band = $400.00 - (2 * $7.00) = $400.00 - $14.00 = $386.00. SPY at $385.67 fell below the $386.00 threshold. This triggered a buy signal for a 2-sigma mean reversion strategy.*
SPY subsequently rallied. By March 29, 2023, SPY closed at $403.65. The 50-day SMA was approximately $400.00. The price returned to the mean. This would trigger a sell signal, capturing the reversion profit.
Not all sigma events lead to mean reversion. Market regimes can shift. During strong trends, prices can stay beyond 2-sigma or 3-sigma for extended periods. This poses a risk for mean reversion strategies. Trend-following strategies benefit in such environments. Mean reversion traders often incorporate regime filters to avoid trading against strong trends. For instance, they might only trade mean reversion when a longer-term moving average indicates a flat or ranging market.
Limitations and Considerations
The assumption of normal distribution for asset prices is a simplification. Real-world returns exhibit "fat tails." This means extreme events (high sigma deviations) occur more frequently than a normal distribution predicts. This characteristic makes sigma events more relevant for mean reversion but also increases the risk of "tail risk" – larger-than-expected losses during prolonged deviations.
Traders must also consider the liquidity of the asset. Highly liquid assets like SPY or AAPL process large orders without significant price impact. Illiquid assets can experience wider spreads and greater price volatility around sigma events, making execution challenging.
Another consideration is the choice of standard deviation type. Simple standard deviation measures historical volatility. Other volatility measures, like Exponentially Weighted Moving Average (EWMA) standard deviation, give more weight to recent price data. This makes them more responsive to current market conditions.
Traders implement stop-loss orders to manage risk when a sigma event fails to revert. For example, a trader might exit a long position if the price falls below -3 sigma after an initial -2 sigma entry. This prevents unlimited losses if the price continues to trend away from the mean.
The profitability of sigma event strategies depends on careful parameter tuning, robust risk management, and understanding market regimes. Backtesting with historical data is essential to validate strategy performance.