Volatility Indicators

Volatility indicators measure how much price moves rather than where it goes: realized measures from past prices (standard deviation, ATR, the Parkinson, Garman-Klass and Yang-Zhang range estimators), implied measures from option prices (implied volatility, IV rank, the volatility risk premium) and the bands that place a volatility-scaled envelope around price. They are read to size positions, set stops, price options and detect regime change.

What do Volatility indicators have in common?

Every indicator in this category estimates the scale of price movement over a window. The simplest is the standard deviation of returns, which uses only closes and so discards most of each day's information. The range-based estimators were developed to use the high and low as well: Parkinson in 1980 from the range alone, Garman and Klass the same year adding the open and close, Rogers and Satchell in 1991 allowing for a drift, and Yang and Zhang in 2000 adding the overnight gap. Each is several times more efficient than the close-to-close estimate on the same data. Average true range, from Wilder, serves the same purpose in price units rather than percentages.

The category divides into three groups. Realized measures (standard deviation, variance, historical, realized, close-to-close, zero-trend close-to-close, OHLC, range-based, Parkinson, Garman-Klass, Rogers-Satchell and Yang-Zhang volatility, realized variance, bipower variation, true range, ATR and its percent and normalized forms, average daily and weekly range, average deviation, the ulcer index, upside and downside volatility) describe what has happened.

Implied measures (implied volatility, IV rank and percentile, the volatility risk premium, volatility of volatility) describe what options are pricing. Envelopes and regime measures (Bollinger bands with %B and bandwidth, Keltner channels, Donchian channels, acceleration bands, Chaikin volatility, the relative volatility index, volatility contraction and expansion) turn a volatility estimate into a level around price or a signal that the regime has changed.

What they share is that volatility clusters: a high reading tends to be followed by a high reading, which is the single most reliable regularity in financial returns and the reason volatility is forecastable where direction is not. They also share a sensitivity to the window and the sampling frequency, since the same day's data gives a different number at daily and at five-minute intervals, and the difference between implied and realized volatility, the volatility risk premium, has been positive on average for decades because option sellers are paid for bearing the risk of its spikes.

How do you choose a Volatility indicator?

• For position sizing and stops, use ATR or a range-based estimator, which capture intraday moves that a close-to-close standard deviation misses
• For comparing today's volatility with history, use IV rank or percentile for options and the ratio of a short window to a long window for realized measures, since a raw level means little without context
• For whether options are expensive, read the volatility risk premium, implied minus realized, and expect it to be positive most of the time
• Prefer Yang-Zhang when the instrument gaps overnight and Garman-Klass when it does not; Parkinson is the fallback when only high and low are available
• Read volatility contraction (a narrow Bollinger bandwidth, a squeeze inside the Keltner channel) as a precondition for a large move, not as its direction

Volatility indicators at a glance

Indicator	What it measures	How it is read
Acceleration Bands	Acceleration Bands, developed by Price Headley, are an envelope around a simple moving average whose width scales with each bar's high-low range relative to its price.	Price pressing against or closing outside the upper band is read as strong momentum; moves back inside the bands suggest the burst is fading.
ATR Percent	ATR percent expresses the Average True Range as a percentage of the closing price, so volatility can be compared across instruments with very different price levels.	A higher percentage means the instrument typically moves more per bar relative to its price.
Average Daily Range	Average Daily Range is the mean of the daily high-low range over a chosen number of days.	It is a simple yardstick for how far price usually moves in a day, used for sizing stops or judging whether today's range is large relative to normal.
Average Deviation	Average deviation, also called mean absolute deviation, is the mean of the absolute distances of values from their mean.	A larger value means observations sit further from their average.
Average Weekly Range	Average Weekly Range is the mean of weekly high-low ranges over a chosen number of weeks.	It gives a baseline for a typical week's movement, so a week ending with a range well above it signals unusually high volatility.
Bipower Variation	Bipower variation estimates the continuous part of price variance from the products of adjacent absolute returns.	Comparing realized variance with bipower variation separates jump risk from diffusive volatility: when realized variance is clearly larger, a jump probably contributed.
Bollinger %B	Percent B shows where the price sits relative to the Bollinger Bands, scaled so that the lower band is 0 and the upper band is 1.	A value of 0.5 is the middle of the bands; above 1 means price is above the upper band, and below 0 means it is below the lower band.
Bollinger BandWidth	Bollinger Bandwidth is the distance between the upper and lower Bollinger Bands divided by the middle band.	Low bandwidth marks a squeeze: volatility is at the low end of its recent range, and periods of low volatility are often followed by expansion, though direction is not implied.
Chaikin Volatility	Chaikin Volatility measures the rate of change of a smoothed high-low range.	Positive values mean the average range is expanding compared with the lookback, negative values mean it is contracting.
Close-to-Close Volatility	Close-to-close volatility is the sample standard deviation of log returns between successive closing prices, usually annualized.	Higher values mean larger typical day-to-day swings.
Downside Volatility	Downside volatility, or downside deviation, measures dispersion using only returns below a minimum acceptable return, usually zero.	It is the denominator of the Sortino ratio and answers how large losses relative to the target have been.
Garman-Klass Volatility	The Garman-Klass estimator measures volatility from open, high, low and close prices of each bar.	Read it like other historical volatility measures: larger values mean wider intrabar ranges.
Normalized Average True Range	Normalized Average True Range (NATR) is the Average True Range expressed as a percentage of the close, as implemented in libraries such as TA-Lib.	Higher values mean larger typical bar-to-bar moves as a share of price, letting you compare instruments with different price scales.
OHLC Volatility	OHLC volatility refers to estimators built from each bar's open, high, low and close rather than closes alone, such as Garman-Klass, Rogers-Satchell and Yang-Zhang.	Higher values mean wider price excursions within the bars.
Parkinson Volatility	The Parkinson estimator measures volatility from the high-low range alone.	Larger values mean wider daily ranges and thus higher volatility.
Range-Based Volatility	Range-based volatility estimates volatility from the high-low range of each bar rather than from closing prices alone.	It uses the fact that a wide range implies large movement, even when the close ends near the open.
Realized Variance	Realized variance is the sum of squared returns over a period, typically built from intraday or daily returns.	It is in squared-return units and adds up over time, so a longer window gives a larger number.
Realized Volatility	Realized volatility is the annualized square root of the realized variance computed from squared returns over a window.	Larger values mean the instrument has been moving more.
Relative Volatility Index	The Relative Volatility Index (RVI), introduced by Donald Dorsey, applies an RSI-style calculation to the standard deviation of price instead of price changes.	Values above 50 mean volatility has been higher on up closes, below 50 that it has been higher on down closes.
Rogers-Satchell Volatility	The Rogers-Satchell estimator measures volatility from open, high, low and close and, unlike Parkinson or Garman-Klass, remains unbiased when the price has a nonzero drift.	Read it like other historical volatility figures: higher means wider price movement within bars.
Standard Deviation	Standard deviation measures how widely a set of values is spread around its mean, in the same units as the data.	A larger value means observations are more dispersed.
True Range	True range, defined by J.	It is the size of a single bar's movement including any gap, in price units.
Ulcer Index	The Ulcer Index, developed by Peter Martin, measures downside risk as the root mean square of percentage drawdowns from the running highest close over a lookback.	A higher value means deeper or longer drawdowns from recent highs.
Upside Volatility	Upside volatility measures dispersion using only returns above a threshold, usually zero.	Comparing upside and downside volatility shows whether variability is skewed toward gains or losses.
Variance	Variance is the average squared deviation of values from their mean.	Larger variance means a wider spread around the mean.
Volatility Contraction	Volatility contraction describes a period when a volatility measure, such as ATR, Bollinger Bandwidth or bar range, falls to a low level relative to its recent history.	A lower ratio means recent movement is quieter than usual.
Volatility Expansion	Volatility expansion describes a period when a volatility measure rises well above its recent norm, such as a bar whose range is much larger than the average range.	A ratio well above 1 marks movement larger than usual, which can accompany news, breakouts or stress.
Volatility of Volatility	Volatility of volatility measures how much a volatility series itself varies, commonly as the standard deviation of changes in volatility or implied volatility.	Higher values mean volatility is unstable: it jumps around instead of drifting.
Volatility Risk Premium	The volatility risk premium is the gap between implied volatility (or implied variance) and the volatility subsequently realized.	On average the premium has been positive for equity indexes, meaning implied volatility has exceeded what was later realized.
Yang-Zhang Volatility	The Yang-Zhang estimator combines overnight (close-to-open) variance, open-to-close variance and the Rogers-Satchell estimator into one measure.	It is read like other historical volatility: higher means more movement.
Zero-Trend Close-to-Close Volatility	Zero-trend close-to-close volatility estimates volatility from squared log returns without subtracting the sample mean, which assumes the average drift over the window is zero.	It matches the usual standard-deviation version when average returns are small, and is a more stable estimate on short windows since it avoids estimating the mean.

Each row is the opening sentence of that indicator's own definition and reading, so the table says exactly what the profile says.

What limits do Volatility indicators share?

Volatility is forecastable but its direction is not, so a contraction says a move is coming without saying which way, and an expansion is usually recognized after the move that caused it. The close-to-close estimators discard most of the day's information and the range-based ones assume continuous trading without gaps or jumps, which is why bipower variation exists to separate the jump component. Implied volatility is the output of a model whose assumptions the market does not believe, IV rank depends on the lookback chosen, and bands scaled by a trailing estimate are widest after the move they would have warned about.

Frequently Asked Questions

What is the difference between historical and implied volatility?

Historical or realized volatility is computed from past prices, usually as the annualized standard deviation of daily returns over a window. Implied volatility is backed out of option prices through a pricing model and represents the volatility the market is charging for the option's life. The gap between them, the volatility risk premium, has been positive on average because option buyers pay for protection against volatility spikes.

Why use Garman-Klass or Yang-Zhang instead of standard deviation?

Because a day's high, low, open and close carry far more information about that day's volatility than its close alone. Garman and Klass showed their estimator was about eight times as efficient as the close-to-close one, meaning it reaches the same precision with an eighth of the data; Yang and Zhang extended it to handle the overnight gap and a drift, which the earlier estimators assumed away.

What does average true range measure?

Average true range is the average over a window, usually 14 periods, of the true range, which is the largest of the day's high minus low, the high minus the previous close and the previous close minus the low. It expresses volatility in price units, so it is used directly to set stops and to size positions so that each carries a similar dollar risk.

What is a volatility squeeze?

A volatility squeeze is a period when a volatility-scaled envelope narrows to an extreme, read most often as Bollinger bandwidth at a multi-month low or the Bollinger bands moving inside the Keltner channel. Because volatility clusters and low volatility tends to be followed by a return to normal, a squeeze is read as the setup for an expansion, with the direction decided by which way price leaves the range.