HammingFilter

Ehlers DSP filter ehlers dsp windowing spectral

Hamming windowed FIR filter with pedestal.

Visual Example

HammingFilter — annotated preview mapping to core implementation

Synthetic ideal per library logic. Generated 2026-07-01 IST via docs/generate_all_previews.py (reproducible; maps to core Next<T> implementation).

Description

Hamming windowed FIR filter with pedestal.

Apply as a windowing function before DFT-based cycle detection to reduce sidelobe leakage and obtain cleaner dominant cycle estimates.

Part of QuantWave's Ehlers digital signal processing suite. Designed for low-lag cycle and trend work — pair with Roofing Filter or SuperSmoother on noisy inputs.

The Hamming window is a raised-cosine weighting function that reduces spectral leakage by tapering the edges of a data block. Ehlers uses it in DFT-based cycle measurement tools to prevent energy in one frequency bin from contaminating adjacent bins, improving cycle period resolution.

Typical applications:

Use for cycle timing in mean-reverting regimes
Gate with Hurst exponent or ADX before taking cycle signals
Allow 20+ bars warm-up for filter state to stabilise
Chain with Roofing Filter when input is noisy

QuantWave implements this via the universal Next<T> trait — bit-identical across Rust streaming, Python streaming, and Polars .ta() batch plugins.

Formula / Specification

Implementation (quantwave-core/src/indicators/hamming.rs):

[ Deg(n) = Pedestal + (180 - 2 \times Pedestal) \times \frac{n}{L-1} ] [ Coef(n) = \sin\left(\frac{Deg(n) \times \pi}{180}\right) ] [ Filt = \frac{\sum_{n=0}^{L-1} Coef(n) \cdot Price_{t-n}}{\sum Coef(n)} ]

Gold-standard parity vectors: quantwave-core/tests/gold_standard/hamming_filter.json.

Parameters

Parameter	Default	Description
`length`	20	Filter length
`pedestal`	10.0	Pedestal in degrees

Usage Examples

Streaming (Rust)

use quantwave_core::indicators::HAMMING_FILTER;
use quantwave_core::traits::Next;

let mut ind = HAMMING_FILTER::new(20);
for price in &prices {
    let value = ind.next(price);
}

Streaming (Python)

from quantwave import HAMMING_FILTER

ind = HAMMING_FILTER(20)
for price in prices:
    value = ind.next(price)

Polars Batch (Python)

import polars as pl
import quantwave as qw

def apply_hammingfilter(series: pl.Series) -> pl.Series:
    ind = qw.HAMMING_FILTER(20)
    return pl.Series([ind.next(float(v)) for v in series.to_list()])

df = (
    pl.read_csv('ohlcv.csv')
    .lazy()
    .with_columns(
        pl.col("close").map_batches(apply_hammingfilter, return_dtype=pl.Float64).alias("hammingfilter")
    )
    .collect()
)

All surfaces are bit-identical via the single Next<T> implementation and proptests.

Edge Cases & Limitations

Recursive DSP filters require a warm-up period; first N bars may be unstable or raw-pass-through.
Designed for cyclic/mean-reverting regimes; trending markets can produce lag or drift.
Parameter period (or equivalent) controls cutoff — too small adds noise, too large adds lag.
Prefer chaining with other Ehlers tools (Roofing Filter, SuperSmoother) on noisy inputs.
Validated via proptests against gold-standard vectors where available.
No look-ahead bias; suitable for live streaming and batch feature pipelines.

Boundary Behavior

Condition	Behavior
Warm-up	Leading bars return NaN until warmup_bars is satisfied.
period > len	When period exceeds series length, output is all NaN.
NaN inputs	NaN in input propagates to output (NaN out).
Invalid params	Non-positive period or missing required params raise ValueError.
Empty data	Empty input returns an empty result series.

Sources & References

Primary Source: https://github.com/lavs9/quantwave/blob/main/references/traderstipsreference/TRADERS’ TIPS - SEPTEMBER 2021.html

Implementation: quantwave-core/src/indicators/hamming.rs (HAMMING_FILTER / HAMMING_FILTER_METADATA). Parity: quantwave-core/tests/gold_standard/hamming_filter.json

Provenance: Standards bulk upgrade 2026-07-01 IST — see docs/DOCUMENTATION_STANDARDS.md.