The Chande Momentum Oscillator (CMO) is an indicator used in technical analysis to measure the strength and direction of recent price changes. Over a chosen number of periods it adds up all the gains and all the losses, takes the difference between the two totals and divides by their sum. The result is multiplied by 100, giving a value between -100 and +100.
The indicator was introduced by the analyst Tushar Chande in the 1990s. It belongs to the same family as the relative strength index and the stochastic oscillator, and is used in similar ways: to identify stretched conditions, to gauge the strength of a trend and to generate entry and exit rules. Its distinguishing feature is that it uses unsmoothed data and puts both gains and losses in the numerator.
Formula and example
The calculation is:
- is the sum of the price changes on periods when the close was higher than the previous close.
- is the sum of the absolute price changes on periods when the close was lower than the previous close.
- Both sums are taken over the last periods. Settings of 9, 14 and 20 are common.
Suppose that over five days a share's closing price changes by +2, -1, +3, -0.5 and +1.5. The gains sum to and the losses to . The oscillator is . Had every day been a gain the reading would have been +100, and had gains and losses been equal it would have been zero.
Interpretation
The sign shows the direction of net movement and the size shows how one-sided it has been. Analysts use the indicator in several ways:
- Overbought and oversold levels. Chande suggested treating readings above +50 as overbought and below -50 as oversold. At +50 the gains over the window are three times the losses.
- Trend strength. The absolute value is read as a measure of how strongly the market is trending, in a way comparable to the average directional index. Values near zero indicate a market moving sideways.
- Signal line. A moving average of the oscillator, often over 9 or 10 periods, is plotted with it, and crossings of the two lines are used as a signal.
- Divergence. A new price high that is not matched by a new high in the oscillator is taken as a sign of fading momentum.
Whether an extreme reading points to continuation or reversal depends on the market. In a strong trend the oscillator can stay above +50 for a long time, while in a range-bound market extremes tend to be followed by mean reversion.
Relationship to the relative strength index
The two indicators use the same raw inputs. The relative strength index divides gains by the sum of gains and losses, which places it on a scale of 0 to 100. If both are computed with simple sums over the same window, they are linked exactly by:
In the example above the relative strength index would be , and . In practice the standard relative strength index applies an exponential smoothing devised by Welles Wilder, so it changes more slowly, and the two diverge. The unsmoothed oscillator reacts faster and reaches extremes more often.
Use in adaptive moving averages
Chande also used the absolute value of the oscillator as a measure of market efficiency in an adaptive average known as the variable index dynamic average. The smoothing factor of an exponential moving average is multiplied by the absolute oscillator reading, expressed as a fraction, so the average follows prices closely when movement is one-directional and flattens when it is not.
Limitations
Because the data are not smoothed, the indicator is noisy at short settings and produces many signals during a period of consolidation. The reading can change sharply when a single large move drops out of the window. Fixed thresholds such as plus or minus 50 are conventions and do not suit every market or time frame.
The indicator is a transformation of past prices and contains no information beyond them. Rules built on it should be evaluated by backtesting on data not used to choose the settings, since selecting the best window and thresholds after the fact is a common route to overfitting. It is normally combined with other evidence, not used alone.