Summary
HilbertPhasor is RTTA's streaming Hilbert in-phase and quadrature components,
porting TA-Lib HT_PHASOR. Together they form a complex analytic signal of the
detrended, smoothed price used inside the dominant-cycle estimator.
Update API
result = rtta.HilbertPhasor(fillna=True).update(value)
| Parameter | Default | Meaning |
|---|---|---|
fillna |
True |
If False, NaN until lookback of 32 samples is exceeded |
update(...) returns a result with:
inphase— in-phase component \(I\)quadrature— quadrature component \(Q\)
advance(value) updates state; last() returns the cached result.
Theory Of Operation
The Hilbert transform produces a \(90^\circ\) phase-shifted companion of a real signal. Ehlers / TA-Lib apply a short FIR-style Hilbert operator to a 4-bar WMA of price (the detrender), then take:
- In-phase as the detrender delayed by three adaptive Hilbert stages
(
i1_for_even_prev3_/i1_for_odd_prev3_depending on bar parity). - Quadrature as the Hilbert transform of the detrender (\(Q1\)).
The complex pair \((I,Q)\) feeds the homodyne discriminator that estimates dominant period. Plotting \(I\) vs \(Q\) traces a phasor diagram of the local cycle.
Recurrence
Shared front end with other Hilbert indicators (see
HilbertDominantCyclePeriod):
Let \(D_t\) be the adaptive Hilbert detrender of \(S_t\), and \(Q1_t\) the adaptive Hilbert transform of \(D_t\). With bar parity alternating odd/even filter banks:
In code after each update:
With fillna=False, both fields are NaN until more than 32 samples have been
processed.
Implementation Notes
- Implemented as
class HilbertPhasorwrappingHilbertCycleEngine::inphase()and::quadrature(). - Result type:
HilbertPhasorResult(inphase,quadrature). - Lookback matches
HT_DCPERIOD/HT_PHASOR(32). - Batch helper:
batch_hilbert_phasor.