On this page
Chande Forecast Oscillator: Price vs Regression Forecast
The **Chande forecast oscillator**, developed by Tushar Chande, compares each closing price to the value that an n-period linear regression line forecasts for the same bar. It plots the percent difference around a zero line and signals whether price is running ahead of, or behind, its statistical trend.
Key Takeaways
- The oscillator equals (Close minus regression forecast) divided by Close, times 100.
- The regression forecast uses the least-squares slope and intercept of the last n closes.
- Readings above zero say price exceeds its trend; below zero says price is undershooting it.
- The most common error is treating the oscillator as bound between fixed extremes.
Key Takeaways
- The oscillator equals (Close minus regression forecast) divided by Close, times 100.
- The regression forecast uses the least-squares slope and intercept of the last n closes.
- Readings above zero say price exceeds its trend; below zero says price is undershooting it.
- The most common error is treating the oscillator as bound between fixed extremes.
What It Is
The Chande forecast oscillator is a deviation oscillator. It does not forecast future prices in the trading sense. It measures how far the current close sits, in percentage terms, from where an n-period linear regression would expect it to be on that same bar.
Tushar Chande, co-author of The New Technical Trader, designed several oscillators that use regression and statistical techniques rather than smoothed averages. The forecast oscillator is part of that family, alongside the Chande momentum oscillator and the variable index dynamic average.
The Intuition
A linear regression line drawn through the last n closes is the best straight-line fit to recent price. If you extend that line to the most recent bar, you get the forecast value, what a straight-line trend says the bar "should" close at.
Comparing the actual close to that forecast tells you whether price is leading, matching, or lagging the local trend. A close well above the forecast says buyers have pushed price above its straight-line path. A close well below says sellers have pulled it under the path. The percentage form makes the signal comparable across assets.
How It Works
The formula is:
Forecast = slope x n + intercept
CFO = ((Close - Forecast) / Close) x 100
Where slope and intercept come from a least-squares fit on the last n closes. Common lookbacks are 14 or 20 bars. The forecast value is the regression line evaluated at the most recent bar, sometimes called the linear regression "endpoint" forecast.
Standard readings:
- CFO above zero: close exceeds the regression forecast; price is above its straight-line trend.
- CFO below zero: close is below the regression forecast; price is under its trend.
- Zero crossings: the close has rejoined the regression line; often used as a trend-shift cue.
- Extreme positive readings: price has stretched well above its regression line; mean-reversion traders treat as exhaustion.
- Extreme negative readings: price has stretched well below its line; sometimes read as oversold.
Chande typically paired the oscillator with a short moving average of itself as a trigger line.
Worked Example
Suppose the last 14 closes have produced a regression slope of 0.30 and intercept of 100. Indexed so that the most recent bar is bar 14, the forecast at bar 14 is:
Forecast = 0.30 x 14 + 100 = 104.20
If today's close is 106, the Chande forecast oscillator reads:
CFO = ((106 - 104.20) / 106) x 100 = 1.70
Price is about 1.7% above its straight-line fit. A trader using the indicator would note that the close has pulled ahead of trend.
If a few bars later the close has dropped to 102 and the regression forecast is now 105.10, the new CFO is:
CFO = ((102 - 105.10) / 102) x 100 = -3.04
A move from +1.70 to -3.04 means price has gone from running ahead of trend to lagging behind it. The zero-cross in between flags the regime shift.
Common Mistakes
- Treating the oscillator as bounded. CFO has no fixed limits. Large readings simply mean a big gap between price and regression, and what counts as "extreme" is asset-specific.
- Ignoring the regression's instability at trend changes. When the underlying trend rotates, the least-squares slope and intercept jump, and the CFO can spike for reasons unrelated to a new signal.
- Using on choppy markets. The regression assumption is a straight-line trend. In choppy ranges the fit is poor and the oscillator is noisy.
- Reading divergence without confirmation. Divergence between price and CFO is meaningful only when paired with structure such as support, resistance, or volume confirmation.
- Mixing up CFO with Chande momentum. Chande published several oscillators. The Chande Momentum Oscillator (CMO) and the Forecast Oscillator share an author but have different formulas and different uses.
Frequently Asked Questions
What is the Chande forecast oscillator in simple terms? The Chande forecast oscillator compares today's close to where a linear regression of recent closes thinks today should be. Readings above zero mean price is running above its trend line; below zero means it is running below.
How does the Chande forecast oscillator affect investment decisions? Trend traders use zero crossings to enter when price retakes its regression trend and to exit when it loses it. Mean-reversion traders look for extreme positive or negative readings as exhaustion signals.
What is a real-world example of the Chande forecast oscillator? On a stock pulling back inside an uptrend, the close often dips below the regression forecast for a few bars, pushing CFO negative, before climbing back above as the trend resumes. The cross back above zero can mark the end of the pullback.
How can investors use the Chande forecast oscillator effectively? Use it on assets that show clean linear trends, pair it with a trend filter, and watch for divergence at structural levels. Avoid reading the oscillator in isolation in choppy markets.
How is the Chande forecast oscillator different from price rate of change? ROC compares today's close to a close from n bars ago. The Chande forecast oscillator compares today's close to a regression-fitted value. ROC is simpler; CFO is more sensitive to the shape of the recent trend.
Sources
- Chande, T. and Kroll, S. The New Technical Trader. Wiley, 1994. https://archive.org/details/newtechnicaltrad0000chan
- Trading Technologies. Chande Forecast Oscillator. https://library.tradingtechnologies.com/trade/chrt-ti-chande-forecast-oscillator.html
- user42 Chart Manual. Forecast Oscillator. https://user42.tuxfamily.org/chart/manual/Forecast-Oscillator.html
- LightningChart. Chande Forecast Oscillator Guide. https://lightningchart.com/blog/trader/chande-forecast-oscillator-indicator/
Disclaimer
This article is educational content only and is not financial advice. Nothing here is a recommendation to buy, sell, or hold any security. Consult a licensed advisor before making investment decisions.