2000issue C091-4
Constructing a Hilbert squelch dominant-cycle gate
A spectral period meter is assembled from mid-price, in-phase and quadrature components, and wraparound-corrected phase increments. A squelch-threshold then marks shorter smoothed periods as trend mode and leaves cycle mode for readings at or above the cutoff.
- Mid-price, the average of each bar high and low, is the default input to in-phase and quadrature extraction.
- In-phase and quadrature components are built from lagged mid-price differences, recursively smoothed, and converted through an arctangent and quadrant rules into a 360-degree phase reading.
- The instantaneous period is the bar count at which clipped phase increments first accumulate to 360 degrees, then optionally smoothed into a published period.
- A default squelch cutoff of 20 flags shorter smoothed periods as trend mode and can be drawn as a highlight or recoded as a 1/0 stream.
Start from a spectral period meter
Spectral analysis, in this construction, means recovering a time-varying cycle length from in-phase and quadrature components of an ordered price series rather than from a fixed lookback oscillator.
The dominant cycle is the bar length implied by accumulating successive phase increments until they complete one full rotation.
The default input to the in-phase and quadrature extraction is mid-price, formed as the average of each bar high and low.
Form a phase angle from in-phase and quadrature
In-phase is the recursively smoothed even-symmetric component taken from lagged mid-price differences.
Quadrature is the recursively smoothed odd-symmetric companion component used with in-phase to form a phase angle.
Those components are converted to a phase angle with an arctangent, and quadrant rules expand that arctangent into a 360-degree phase reading.
Accumulate clipped increments to a period
Successive phase changes are wraparound-corrected and then clipped so each increment stays between 1 and 60 degrees.
The instantaneous period is the lookback count at which clipped phase increments first sum to a full circle. In the archive workflow that count is the bar count at which the increments first accumulate to 360 degrees, searched over a window of up to 40 increments, and it is taken before optional smoothing.
A missing instantaneous estimate is replaced by the prior value, and the published period is 0.25 times the new estimate plus 0.75 times the previous smoothed period.
Install the squelch-threshold
A default squelch cutoff of 20 marks bars whose smoothed period is shorter than that cutoff.
The squelch-threshold is the minimum accepted period. Readings shorter than this cutoff are treated as a separate regime from cycle mode.
Implementations use that period-versus-cutoff test to label bars as trend mode or cycle mode. Trend mode is the regime flag used when the smoothed period is below the squelch cutoff. Cycle mode is the regime flag used when the smoothed period is at or above the squelch cutoff.
Feed a display or a later model input
The below-cutoff condition is rendered as a chart highlight or recoded as a 1/0 stream so the same construction can feed a display or a later model input.
All readings on this track · 21 readings
- 1982Spectrum-first construction of moving-average trend filters
- 1990Crowd questions that switch cycle regimes
- 1993Antithetic twins to cancel origin-linked forecast bias
- 1994Constructing frequency response from a three-stage filter cascade
- 1994Bench-testing smoothers with cycle phase and power
- 1995Constructing minimum tradable cycles from spectral noise
- 1995Spectral whipsaw limits for a price-crossing average
- 1996Constructing market phase from dominant-cycle components
- 2000Constructing a Hilbert squelch dominant-cycle gate
- 2007When cycle mode fails: regime gates for spectral trades
- 2008Constructing singular-spectrum forecast baselines
- 2008Stacking a midband cycle gate on a short-memory trend filter
- 2011Engineering spectral cycle detection and moving-average harmony
- 2011Evaluating fixed oscillators against a moving dominant cycle
- 2011A construction checklist for a dominant-cycle sinewave
- 2012Assembling a market-mode heatmap from successive band-pass filters
- 2012Adaptive trend detection from dominant-cycle spectra
- 2015Constructing price paths from cycles and support
- 2018Oscillator distribution shape for mean-reversion filters
- 2020Constructing a cosine-correlation cycle and phasor trend filter
- 2020Constructing a correlation-cycle market-state engine