Direct Answer
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. It still does not account for opening gaps. Read it like other historical volatility figures: higher means wider price movement within bars.
Rogers-Satchell Volatility: Formula, Meaning, Signals, Examples and How to Use It
How Is Rogers-Satchell Volatility Calculated?
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. It still does not account for opening gaps.
Formula: sigma^2 = (1 / n) x sum of [ ln(High/Close) x ln(High/Open) + ln(Low/Close) x ln(Low/Open) ]; annualize with sqrt(252)
Inputs:
- Open, high, low and close for each bar
- Window length n
- Annualization factor
Worked example. For a single bar with open 99, high 102, low 98 and close 101, the per-bar volatility is 2.45%. The inputs are illustrative numbers, not market data.
Reading it. Read it like other historical volatility figures: higher means wider price movement within bars. Its main advantage is that a trending market does not inflate it the way it can distort other range estimators.
Limits. It ignores the jump between one close and the next open, so it underestimates volatility for assets that gap a lot. Single-bar values are noisy and the estimator is meant to be averaged over a window.
What is 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. It still does not account for opening gaps.
How to use Rogers-Satchell Volatility
Read it like other historical volatility figures: higher means wider price movement within bars. Its main advantage is that a trending market does not inflate it the way it can distort other range estimators. A reading alone is never a buy or sell signal; it needs context from price structure, volume, volatility and the broader market regime.
Strengths
- Each bar's term is a sum of two products of log distances, so it can be computed from four prices with no model parameters.
- Unlike range-only formulas, it uses where the open and close sit inside the high-low range, so trending bars are treated differently from ranging bars.
- Reproducible when the formula and data convention are fixed.
Weaknesses and limitations
- Opening at the low and closing at the high gives the bar a term of exactly zero, because each product contains a zero log distance.
- Discrete sampling puts the observed high and low inside the true extremes, so the log distances are slightly small and the estimate leans low.
- Correlated indicators can create false confidence by repeating the same underlying information.
Editorial and risk note
This page is educational content, not individualized investment advice. It should not imply guaranteed prediction, accuracy or outperformance.
Frequently Asked Questions
Is Rogers-Satchell Volatility a buy or sell signal?
No. Read it like other historical volatility figures: higher means wider price movement within bars. It ignores the jump between one close and the next open, so it underestimates volatility for assets that gap a lot. A trade built on it still needs a trigger, an invalidation rule, a position size and realistic execution assumptions.
What is the best setting for Rogers-Satchell Volatility?
Rogers-Satchell Volatility takes these inputs: open, high, low and close for each bar; window length n; annualization factor. There is no universal best setting for the tunable ones, so test them across instruments, regimes and out-of-sample periods and favor stable parameter regions over one historical winner.
Can Rogers-Satchell Volatility be used by itself?
It ignores the jump between one close and the next open, so it underestimates volatility for assets that gap a lot. Single-bar values are noisy and the estimator is meant to be averaged over a window. For that reason it is read next to measures of direction, participation, valuation, liquidity and risk, not alone.
Does Rogers-Satchell Volatility work on every timeframe?
The calculation (sigma^2 = (1 / n) x sum of [ ln(High/Close) x ln(High/Open) + ln(Low/Close) x ln(Low/Open) ]; annualize with sqrt(252)) can be run on any bar size, but behavior is not portable: a reading on a five-minute chart describes a very different market window from the same reading on a daily chart. Validate the exact timeframe before relying on it.
Why does Rogers-Satchell Volatility give false signals?
Opening at the low and closing at the high gives the bar a term of exactly zero, because each product contains a zero log distance. Discrete sampling puts the observed high and low inside the true extremes, so the log distances are slightly small and the estimate leans low. Lag, noise and regime changes add to this, because market participants react to new information after the reading is calculated.
How should Rogers-Satchell Volatility be backtested?
Rebuild Rogers-Satchell Volatility from open, high, low and close for each bar using point-in-time data, write the rules before testing, include delisted securities where relevant, model realistic fills and costs, reserve a validation sample and compare with a simpler baseline.