How it works
The Relative Volatility Index takes the arithmetic of RSI and points it at a different input. Where RSI compares the size of up moves with the size of down moves, RVI compares the amount of volatility that occurs on up bars with the amount that occurs on down bars. The result is a bounded oscillator that answers a subtly different question: not which direction has moved further, but which direction is being traded with more force.
The intuition is that markets rarely rise and fall with the same character. Advances in equity indices are typically slow and low-volatility, while declines are fast and disorderly, and a measure that separates volatility by direction picks that asymmetry up. When an advance starts to run with expanding volatility, the buying has become urgent rather than passive, and that is a different regime from a quiet drift higher even if the price change is identical.
It was designed as a confirming indicator rather than a standalone one, and that is still the honest way to use it. Its author suggested pairing it with a trend or momentum tool and only taking signals where the two agree, which filters a large share of the whipsaws that any oscillator produces on its own. A moving average crossover confirmed by RVI above fifty is a materially different proposition from the same crossover with RVI at thirty, because in the second case the move is happening on shrinking volatility, which is characteristic of a drift that fades.
Because it is built from the standard deviation rather than from price change, RVI does not track price the way RSI does, and comparing the two side by side is instructive. RSI can be pinned high in a slow grinding advance where RVI stays middling because nothing forceful is happening. Conversely a violent two-day shakeout inside an uptrend can drop RVI sharply while RSI barely notices. Traders who find RSI too correlated with price often prefer RVI for exactly that reason.
Calculation
The arithmetic in words, in the order it happens.
For every bar compute the standard deviation of the close over a short window, conventionally 10 periods. Then classify each bar by direction: if the close is higher than the previous close, that bar's standard deviation goes into the up bucket and zero goes into the down bucket; if the close is lower, the reverse. Smooth each bucket with Wilder's method over a longer window, conventionally 14 periods, exactly as RSI smooths average gain and average loss. The Relative Volatility Index is 100 times the smoothed up value divided by the sum of the smoothed up and down values, giving a series bounded between 0 and 100. A later revision by the same author computes the measure separately on the high and on the low and averages the two results.
Source
An AlgoBeamScript implementation of the formula above, written by us from the arithmetic so the code and the calculation agree line for line.
Runs unchanged on the platform and in the AlgoBeamTS runtime. The language reference is in the documentation.
Inputs
Defaults are the values most charting packages ship with. They are conventions, not optimal settings — the right length depends on your instrument and your holding period.
| Input | Default | What it changes |
|---|---|---|
| Standard deviation length | 10 | Window used to measure volatility on each bar before it is sorted by direction. Shorter values make the oscillator react to individual violent bars; longer values describe the character of a whole swing. |
| Smoothing length | 14 | Wilder smoothing applied to the up and down volatility buckets. Raising it produces a slow line that only changes character at genuine regime shifts; lowering it produces frequent crossings of the midline. |
| Source | Close | The series whose dispersion is measured. The revised version of the indicator averages results computed from the high and the low, which reduces sensitivity to erratic closing prints. |
How to read it
What practitioners take from the plot. Read these as descriptions of market state, not as entry signals.
- Above 50
- More volatility is occurring on advancing bars than on declining ones. Used as a confirmation filter for long entries generated elsewhere.
- Below 50
- Declines are the forceful side of the market. Long signals from other tools are questionable here even if price is still rising.
- Above 60 while a trend signal fires
- The strong confirmation the indicator was designed to give: direction and force agree, which historically filters a meaningful share of failed breakouts.
- Price making new highs while RVI falls
- The advance is proceeding on shrinking volatility. Not a reversal signal, but a warning that the move is drifting rather than being driven.
Limitations
Where this indicator misleads. None of these are fixed by a better parameter.
- It is a confirming tool. Used alone it produces the same whipsaw problem as any bounded oscillator, and its author was explicit that it was never intended as a standalone entry trigger.
- The standard deviation input is computed over a short window, so RVI is jumpy around single outlier bars and can flip across the midline on one violent session.
- In very quiet markets the volatility being sorted is tiny in absolute terms, and the oscillator still swings across its full range, giving confident-looking readings about differences that are economically meaningless.
- It is not widely used, which means fewer eyes and less collective knowledge about how it behaves in unusual regimes than for RSI or ATR.
Educational reference. This page explains how an indicator is built and how it is commonly read. It is not investment advice, not a recommendation and not a signal service. No indicator is profitable on its own — each is a way of describing a market, and any rule built on one has to be tested with realistic costs before it is traded.