Skip to main content
Track Spectral analysis
6 / 21
Library

1995issue C091-6

Constructing minimum tradable cycles from spectral noise

Before cycles are hunted, the archive requires a noise measure because a simple moving average is said to need cyclic variation. Bar-range uncertainty is turned into a spectral magnitude floor, and only the first peak above that floor is used to size a moving-average crossover.

  • A simple moving average is presented as a basic above-or-below reading of a price series, and it is said to need cyclic variation, so a noise measure is required before cycles are hunted.
  • Average-range noise and a chosen signal-to-noise ratio set a cycle cutoff, because a crossover fill is assumed to occur somewhere inside the high-low range rather than at a known price.
  • The minimum tradable cycle is the first spectral peak above that cutoff. Representative spectra increase in magnitude toward longer wavelengths, and many short cycles are treated as small-magnitude noise.
  • After that wavelength is selected, the moving-average crossover period is often set to half the wavelength, or slightly less. The cutoff is a low-pass rule that can admit longer peaks but not shorter ones.
Entries in this reading3 entries

Why a noise measure comes first

A simple moving average is presented as a basic reading of a price series. Values above the average are treated as an advance and values below as a decline.

Before cycles are hunted in a price series, a noise measure is required, because a simple moving average is said to need cyclic variation in order to operate.

How cycle energy is located

Fourier-family transforms and maximum-entropy spectral analysis are listed as ways to locate cycle energy. A discrete or fast Fourier transform produces a spectral plot whose peaks mark where sine-cosine energy concentrates.

Representative spectra are described as increasing in magnitude toward longer wavelengths, with many short cycles treated as small-magnitude noise.

Turning bar-range uncertainty into a floor

Average-range noise is a smoothed high-minus-low range used as a bar-level estimate of unobserved intra-period path uncertainty. One illustrated daily series used a 75-bar lookback.

The path inside a bar is treated as unobserved, so a crossover fill is assumed to occur somewhere inside the high-low range rather than at a known price.

A theoretical lower signal-to-noise ratio of one half is derived by assuming worst-case fills at the noise extremes around a unit-magnitude sine wave whose peak-to-peak span is 2. A practical construction floor near 2 is then proposed as about four times that theoretical minimum.

The first peak above the cycle cutoff

The minimum tradable cycle is the first spectral peak above a cycle cutoff equal to the chosen signal-to-noise ratio times the measured noise level. With noise 0.381 and a ratio of 2 the cutoff is 0.762, and the illustrated first peak is 207.7 trading days at magnitude 1.032.

Shorter peaks are treated as below the noise floor. The cutoff is described as a low-pass rule that can admit longer peaks but not shorter ones.

Sizing the moving-average crossover after the cutoff

After a tradable wavelength is selected, a moving-average crossover period is often set to half that wavelength or slightly less.

TradersWeek editorial note: apply that half-wavelength habit only after the cutoff gate has already dropped shorter peaks. The archive states the length choice and the low-pass rule. It does not describe the sequence as a two-gate design.

Educational research material, not investment advice. Historical source context does not establish present-day performance.
6 of 21 in the Spectral analysis track
19951-10 pp.Next on Spectral analysisSpectral whipsaw limits for a price-crossing averageOnce a tradable cycle has already set the simple-moving-average length, the remaining modeled hurdle is whether a shorter residual cycle can displace the price-path crossover times.
All readings on this track · 21 readings
  1. 1982Spectrum-first construction of moving-average trend filters
  2. 1990Crowd questions that switch cycle regimes
  3. 1993Antithetic twins to cancel origin-linked forecast bias
  4. 1994Constructing frequency response from a three-stage filter cascade
  5. 1994Bench-testing smoothers with cycle phase and power
  6. 1995Constructing minimum tradable cycles from spectral noise
  7. 1995Spectral whipsaw limits for a price-crossing average
  8. 1996Constructing market phase from dominant-cycle components
  9. 2000Constructing a Hilbert squelch dominant-cycle gate
  10. 2007When cycle mode fails: regime gates for spectral trades
  11. 2008Constructing singular-spectrum forecast baselines
  12. 2008Stacking a midband cycle gate on a short-memory trend filter
  13. 2011Engineering spectral cycle detection and moving-average harmony
  14. 2011Evaluating fixed oscillators against a moving dominant cycle
  15. 2011A construction checklist for a dominant-cycle sinewave
  16. 2012Assembling a market-mode heatmap from successive band-pass filters
  17. 2012Adaptive trend detection from dominant-cycle spectra
  18. 2015Constructing price paths from cycles and support
  19. 2018Oscillator distribution shape for mean-reversion filters
  20. 2020Constructing a cosine-correlation cycle and phasor trend filter
  21. 2020Constructing a correlation-cycle market-state engine
All 63 readings tagged Spectral analysis
Also on Spectral analysis5 readings