Volatility is the enemy of compounding. High volatility portfolios require deep drawdowns to recover. A 50% loss requires a 100% gain to break even. Reducing volatility extends investment horizons and improves risk-adjusted returns. Volatility management is a core discipline in systematic investing.
Measuring Volatility
Standard deviation quantifies volatility. Daily returns' standard deviation annualized gives annualized volatility. But volatility changes over time. Realized volatility from yesterday's data differs from tomorrow's volatility. Adaptive measures account for this non-stationarity.
GARCH models estimate volatility dynamically, updating daily based on recent price movements. Implied volatility from option prices reveals market expectations. Using both realized and implied volatility provides fuller picture of portfolio risk.
Volatility Targeting
Target volatility strategies adjust leverage based on realized volatility. When volatility is low, increase leverage. When volatility spikes, reduce leverage. This keeps portfolio volatility constant despite underlying market shifts. Constant volatility portfolios compound more smoothly.
Dynamic position sizing achieves the same goal. Position sizes inversely proportional to underlying volatility. High volatility stocks receive smaller allocations; low volatility stocks receive larger allocations. This balances contribution to portfolio risk.
Volatility Clustering
Volatility is not random. High volatility periods cluster together (crises), as do low volatility periods (calm markets). This clustering means volatility is somewhat predictable. Strategies that adjust to volatility regimes outperform static allocation strategies.
Regime detection is critical. Is the market in a low-volatility trending environment or a high-volatility mean-reverting environment? Different strategies work in different regimes. Adaptive portfolios detect regime changes and adjust accordingly.
Tail Risk and Black Swans
Standard deviation ignores tail events. A portfolio with normal volatility can suffer 10x normal-volatility losses in crises. Black swan events happen; they're more common than statistical models suggest. Portfolios should assume tails are fatter than normal distributions predict.
Stress testing and scenario analysis prevent tail surprises. Assume correlations go to 1 in crises. Assume liquidity dries up. Model portfolio performance in 2008-like scenarios. Knowing maximum possible losses prevents blindsided destruction.
Educational content only. Not investment advice.