Skip to main content
Track Exponential smoothing
31 / 41
Search the archiveLibrary
Library

2004issue C041-10

Piecewise trend quality from moving-average reversals

Trend quality is assembled in order. A moving-average reversal-gate is allowed to reset state, a piecewise-exponential-smooth must not leak across that reset, and only then are residual-noise, the quality-reading, and the balance-reading formed so each term can be checked.

  • The reversal-gate is a signed flag from a fast and a slow moving average of the close. A change in that flag starts a new trend segment.
  • Cumulative-price-change adds signed close-to-close increments only inside the current segment and is cleared to zero when the gate flips.
  • The piecewise-exponential-smooth of that accumulator is restarted at each crossover so one segment cannot bleed into the next. Residual-noise is then taken from the gap and scaled.
  • The quality-reading is the piecewise trend divided by residual-noise when that noise estimate is positive. The balance-reading is the absolute piecewise trend as a share of absolute trend plus noise, usually scaled from 0 to 100.
Entries in this reading3 entries
Current trackExponential smoothing183 readingsTrend filter184 readingsMoving average597 readings

Build the trend-filter in order

A trend-filter here is the assembled construction that converts ordered closes into a segmented strength and balance forecast rather than a raw price average.

The archive workflow builds that object in pieces. First a moving-average reversal-gate is allowed to reset state. Then a piecewise-exponential-smooth is applied to a segment accumulator and must not leak across that reset. Only afterward is a trend-to-noise ratio formed.

Start with a reversal-gate

A fast and a slow moving average of the close create a signed reversal flag. A change in that flag marks the start of a new trend segment.

Worked reconstructions pair a 7-period fast exponential average with a 15-period slow exponential average as the default reversal-gate.

Keep cumulative-price-change inside one segment

Inside a segment, cumulative-price-change adds each signed close-to-close increment. When the reversal flag changes, that accumulator is reset to zero.

Some reconstructions keep a separate up-segment accumulator and down-segment accumulator and then join them, so a directional restart cannot carry the tail of the opposite run.

Restart the piecewise-exponential-smooth

The trend term is a piecewise-exponential-smooth of the segment accumulator, commonly using a 4-period span. The smoother is restarted at each crossover so one segment cannot bleed into the next.

Form residual-noise from the gap

Residual-noise is taken from the gap between the segment accumulator and the piecewise trend, then scaled by a correction factor. The default noise lookback is 250, and 2 is a common scale.

The noise term is not unique. One family uses a root-mean-square of squared residuals. Another uses a simple average of absolute residuals. Both are then multiplied by the same style of correction factor.

Only then form the quality-reading and balance-reading

The quality-reading is the piecewise trend divided by the noise estimate whenever that noise estimate is positive.

The companion balance-reading is the absolute piecewise trend divided by the sum of absolute trend and noise, then scaled by 100 when that denominator is positive.

CTXS daily Q-indicator after 7/15 average reversals

The quality-reading stays elevated only inside one moving-average segment and falls back to the zero line when the 7-period and 15-period exponential averages cross. Heights were read from the red Q-indicator pane on the Citrix daily vendor chart; the source did not print a table of these values.
The quality-reading stays elevated only inside one moving-average segment and falls back to the zero line when the 7-period and 15-period exponential averages cross. Heights were read from the red Q-indicator pane on the Citrix daily vendor chart; the source did not print a table of these values.CTXS · Daily · 2003-02-01T00:00:00.000Z to 2004-02-29T00:00:00.000Z

Pane label is QIndicator(Close,7,15,4,250,2). Dates follow the February 2003–February 2004 month ticks. Readings are to the nearest half-unit; the raster does not support finer precision.

Screen after the series exist

Screening bands treat quality-reading magnitudes of 1, 2, and 5, and the matching negatives, as weak, moderate, and strong classes. They treat balance-reading values of 50, 65, and 80 as successively stronger trend-dominant states.

Educational research material, not investment advice. Historical source context does not establish present-day performance.
31 of 41 in the Exponential smoothing track
20041-2 pp.Next on Exponential smoothingNested smoothing recipes versus unaudited oscillator experimentsNested-exponential-smoothing first applies a 13-interval exponential moving average to closes and only then a 21-interval rate-of-change-on-smoother, which can be written as the difference or the ratio of the current 13-interval average and its value 21 intervals earlier.
All readings on this track · 41 readings
  1. 1984Matching average age across price filters
  2. 1984Residual-impact construction of exponential smoothers
  3. 1988Constructing simple and exponential moving averages as specified filters
  4. 1989Evaluating smoothed free reserves and funds-discount spreads
  5. 1990Integer construction of moving averages and exponential smoothing
  6. 1990Constructing incremental moving average filters
  7. 1991Constructing a double-smoothed stochastic oscillator
  8. 1991Constructing an exponential average from a smoothing weight
  9. 1991Constructing a double-smoothed momentum oscillator family
  10. 1991Competing certainty rules from exponential-average conditions
  11. 1991Hybrid construction of a mechanical timing system from smoothed signals
  12. 1991Constructing an exponential smoother from a moving-average length
  13. 1991Equivalent relative-strength-index constructions from a ratio and from smoothed momentum
  14. 1991Constructing the true strength index from double-smoothed momentum
  15. 1991True strength index from signed, double-smoothed momentum
  16. 1992Constructing exponential averages from simple lookbacks
  17. 1992Constructing exponential averages by the smoothing constant, startup length and lag
  18. 1992Nested range smoothing and windowed high-low damping
  19. 1992A weekly equity stance from the discount-bill spread
  20. 1993Double-smoothed stochastics momentum index construction
  21. 1993Constructing the true strength index with double exponential smoothing
  22. 1994Polarized fractal efficiency with a shared exponential smoother
  23. 1994Five-interval exponential smoothing of ten-interval signed path-efficiency
  24. 1994Hand-ranked price-volume series with stacked smoothing and stochastic
  25. 1995Range expansion filter on multi-horizon averages
  26. 1995The exponential smoothing constant from averaging lag
  27. 1997Construct a nested-average rainbow band and oscillator
  28. 1998Filter stacks that combine a moving average with exponential smoothing
  29. 2002Smoothed relative-strength construction from wilder-averaging and a price prefilter
  30. 2003Reverse-engineered RSI price targets
  31. 2004Piecewise trend quality from moving-average reversals
  32. 2004Nested smoothing recipes versus unaudited oscillator experiments
  33. 2008Weight construction beyond the N-day label
  34. 2010A consensus filter for objective RSI and smoothing construction
  35. 2012Sentiment-zone oscillator from signed closes
  36. 2012Classic RSI one-over-N smoothing versus the standard exponential average
  37. 2013Why exponential smoothing uses two over n plus one
  38. 2015Constructing ratio indicators with moving-average and exponential-smoothing weights
  39. 2015Slow relative-strength index from an exponentially smoothed price
  40. 2017Rebuilding the relative strength index as open-scale buy-sell pressure
  41. 2017Comparing Wilder and exponential smoothing inside a relative-strength index
By fieldIndicators and filters874 readingsBy year2004180 readings
All 183 readings tagged Exponential smoothing
Also on Exponential smoothing5 readings
1984Matching average age across price filters1995Range expansion filter on multi-horizon averages1997Construct a nested-average rainbow band and oscillator2012Sentiment-zone oscillator from signed closes1989Alpha-beta lag parameters versus moving-average windows