Volatility
Volatility measures how much an asset's price fluctuates over time, usually expressed as the annualized standard deviation of returns; higher volatility means larger and less predictable price swings.
Volatility quantifies the dispersion of an asset's returns around their average. In practice it is most often calculated as the standard deviation of periodic (daily, weekly, or monthly) returns, then annualized by multiplying by the square root of the number of periods in a year (for example, multiplying daily volatility by the square root of roughly 252 trading days). A stock with 15% annualized volatility is expected, in a typical year, to move within a wider band than one at 8%. Crucially, volatility is symmetric: it treats upside and downside moves equally, which is why some investors prefer downside-only measures.
Volatility is the most common proxy for risk in modern finance. It feeds directly into the Sharpe ratio (which divides excess return by volatility), into the Capital Asset Pricing Model through beta, and into options pricing, where expected future volatility is the single most important unobservable input. Volatility is not constant: it clusters in time (calm periods follow calm periods, turbulent ones follow turbulence), tends to spike during market stress, and often rises faster than it falls. Understanding it helps investors size positions, set expectations for drawdowns, and avoid mistaking a lucky calm stretch for genuine safety.
There is an important distinction between realized (historical) volatility, computed from past prices, and implied volatility, which is inferred from current option prices and reflects the market's forward-looking expectation. Realized volatility is backward-looking and concrete; implied volatility is a forecast embedded in what traders are willing to pay for protection. Both matter, and they frequently diverge, which itself is a tradable signal.
On hedgewing.ai, volatility is woven through the platform's risk analytics rather than treated as an afterthought. It is the denominator behind the Sharpe ratio, a building block of Value at Risk estimates, and one of the engineered features the four-model deep-learning ensemble (LSTM, GRU, TCN, and Transformer) consumes when forming predictions, since recent volatility regime is highly informative about near-term price behavior. The platform's nightly walk-forward backtesting also reports return statistics alongside their volatility so that performance is always shown on a risk-adjusted basis, not as raw return in isolation.
Related terms
Standard Deviation · Implied Volatility · Sharpe Ratio · Value at Risk (VaR)
Back to the hedgewing.ai glossary · See AI stock forecasts