Files
wickra/crates/wickra-core/src/indicators/evwma.rs
T
kingchencandGitHub 466faddd87 feat: Family 01 Moving Averages — ALMA / McGinley / FRAMA / VIDYA / JMA / Alligator / EVWMA (#39)
* feat(alma): add Arnaud Legoux Moving Average

Gaussian-weighted moving average with configurable centre (offset in
[0, 1]) and kernel width (sigma > 0). Pre-computes normalised weights
at construction so each update is a single rolling window dot product.

Reference: Arnaud Legoux and Dimitrios Kouzis-Loukas, 2009.

Touchpoints:
- crates/wickra-core: alma.rs + mod.rs + lib.rs re-export
- bindings/python: PyAlma + __init__.py + test_new_indicators +
  test_known_values reference
- bindings/node: AlmaNode + index.d.ts/index.js + indicators.test.js
  factory + reference value
- bindings/wasm: wasm_scalar_indicator! macro
- fuzz: indicator_update target covers ALMA(9, 0.85, 6.0)
- crates/wickra/benches: bench_scalar entry
- README + CHANGELOG: Moving Averages row + Unreleased entry

* feat(mcginley): add McGinley Dynamic moving average

John McGinley's self-adjusting moving average with the recurrence
MD + (price - MD) / (0.6 * period * (price / MD)^4). Speeds up when
price falls below the indicator and damps when price runs above the
indicator. Seeded with the simple average of the first period inputs.

Reference: McGinley, Technical Analysis of Stocks & Commodities, 1990.

Touchpoints:
- crates/wickra-core: mcginley_dynamic.rs + mod.rs + lib.rs re-export
- bindings/python: PyMcGinleyDynamic + __init__.py + test_new_indicators
  + test_known_values reference
- bindings/node: McGinleyDynamicNode (scalar macro) + index.d.ts/index.js
  + indicators.test.js factory + reference value
- bindings/wasm: wasm_scalar_indicator! macro
- fuzz: indicator_update target covers McGinleyDynamic(10)
- crates/wickra/benches: bench_scalar entry
- README + CHANGELOG: Moving Averages row + Unreleased entry

* feat(frama): add Fractal Adaptive Moving Average

Ehlers' FRAMA adapts its smoothing constant to the fractal dimension of
the recent window: tight tracking in trends, heavy smoothing in chop.
Uses the close-only variant where max/min over each window half drive
the dimension estimate. Period must be even (default 16).

Reference: Ehlers, Fractal Adaptive Moving Average, 2005.

Touchpoints:
- crates/wickra-core: frama.rs + mod.rs + lib.rs re-export
- bindings/python: PyFrama + __init__.py + test_new_indicators +
  test_known_values reference (constant series + uptrend tracking)
- bindings/node: FramaNode (scalar macro) + index.d.ts/index.js +
  indicators.test.js factory + reference value
- bindings/wasm: wasm_scalar_indicator! macro
- fuzz: indicator_update target covers Frama(16)
- crates/wickra/benches: bench_scalar entry
- README + CHANGELOG: Moving Averages row + Unreleased entry

* feat(vidya): add Variable Index Dynamic Average

Chande's VIDYA — an EMA whose alpha scales with |CMO(cmo_period)| / 100.
Strong directional momentum lifts the smoothing constant toward the
EMA-of-period rate; flat or choppy windows shrink it toward zero so
VIDYA coasts on its previous value. Two parameters: period (14) and
cmo_period (9). Reuses the existing wickra-core Cmo internally.

Reference: Chande, Stocks & Commodities, 1992.

Also fixes a silent gap from d37fbd1 (feat(frama)): the PyFrama Python
class wrapper and its add_class registration were dropped because the
two edits hit "File has not been read yet" errors that scrolled past
in a batch. Adds them here alongside VIDYA's bindings.

Touchpoints (VIDYA): vidya.rs + mod.rs + lib.rs re-export, PyVidya +
__init__.py + test_new_indicators + test_known_values reference,
VidyaNode (manual two-param binding) + index.d.ts/index.js +
indicators.test.js factory + reference, wasm_scalar_indicator! macro,
fuzz target, bench, README + CHANGELOG.

* feat(jma): add Jurik Moving Average

Three-stage filter reconstruction of Mark Jurik's adaptive MA (the
algorithm is proprietary; this is the form used by most open-source
ports since the 1999 TASC article). Parameters: period (14), phase in
[-100, 100] (0), power in 1..=4 (2). State is seeded by setting
e0 = JMA = first input so a constant input stream is reproduced exactly.

Touchpoints: jma.rs + mod.rs + lib.rs re-export, PyJma + __init__.py +
test_new_indicators + test_known_values reference, JmaNode (manual
three-param binding) + index.d.ts/index.js + indicators.test.js factory
+ reference, wasm_scalar_indicator! macro, fuzz target, bench, README +
CHANGELOG.

* feat(alligator): add Bill Williams Alligator

Three SMMA lines (Jaw / Teeth / Lips) over the median price
(high + low) / 2 with default periods 13 / 8 / 5. Multi-output
indicator returning AlligatorOutput { jaw, teeth, lips }. The
original chart variant shifts each line forward for display; we
publish the unshifted SMMA values and leave the visual shift to
the consumer.

Reference: Bill Williams, Trading Chaos, 1995.

Touchpoints: alligator.rs + mod.rs + lib.rs re-export, PyAlligator
(Candle input, returns 3-tuple, ndarray (n, 3) batch) + __init__.py
+ test_new_indicators + test_known_values reference, AlligatorNode +
AlligatorValue + index.d.ts/index.js + indicators.test.js multi
factory + reference, WasmAlligator (manual JsValue object) +
candle-fuzz target + README + CHANGELOG.

* feat(evwma): add Elastic Volume-Weighted Moving Average

Christian P. Fries' elastic recurrence where the smoothing weight is the
bar's volume relative to the running window total:

  V_sum_t = sum of volumes over the last period candles
  EVWMA_t = ((V_sum_t - v_t) * EVWMA_{t-1} + v_t * close_t) / V_sum_t

A bar whose volume is small barely moves the average; a bar that
dominates the window pulls it strongly toward that bar's close. Seeded
with the close of the first full window; holds its previous value if
the entire window has zero volume.

Reference: Fries, Wilmott Magazine, 2001.

Touchpoints: evwma.rs + mod.rs + lib.rs re-export, PyEvwma (close +
volume batch) + __init__.py + test_new_indicators CANDLE_SCALAR +
test_known_values reference, EvwmaNode + index.d.ts/index.js +
indicators.test.js candleScalar factory + reference, WasmEvwma,
candle-fuzz target + README + CHANGELOG.

* ci: Force local wheel install in Python jobs

Use --no-index --no-deps so the Python matrix installs the freshly
built wheel from dist/ and never falls back to PyPI. Previously pip
sometimes picked the released 0.2.x wheel on macOS / Windows when its
platform tag was a wider match than the local build, which made the
job test the released package and miss any new symbols added in the
PR (e.g. AttributeError: module 'wickra' has no attribute 'ALMA').
numpy is already installed by the preceding pip step, so --no-deps
is safe.
2026-05-25 15:01:14 +02:00

239 lines
7.3 KiB
Rust

//! Elastic Volume-Weighted Moving Average (EVWMA).
use std::collections::VecDeque;
use crate::error::{Error, Result};
use crate::ohlcv::Candle;
use crate::traits::Indicator;
/// Christian P. Fries' Elastic Volume-Weighted Moving Average.
///
/// Unlike `VWMA` which is a per-bar weighted mean, `EVWMA` runs an
/// "elastic" recurrence whose smoothing weight is the bar's volume relative
/// to the running window-volume:
///
/// ```text
/// V_sum_t = Σ volume_i over the last `period` candles
/// EVWMA_t = ((V_sum_t - volume_t) * EVWMA_{t-1} + volume_t * close_t) / V_sum_t
/// ```
///
/// A bar whose volume is small compared to the window total barely moves the
/// average; a bar whose volume dominates the window pulls it strongly toward
/// the bar's close. The series is seeded with the close of the first candle
/// after the volume window has filled (i.e. after `period` candles).
///
/// If `V_sum_t == 0` (every candle in the window has zero volume), the
/// recurrence is undefined; the indicator holds its previous value.
///
/// Reference: Christian P. Fries, *Wilmott Magazine*, 2001.
///
/// # Example
///
/// ```
/// use wickra_core::{Candle, Evwma, Indicator};
///
/// let mut evwma = Evwma::new(20).unwrap();
/// let mut last = None;
/// for i in 0..40 {
/// let p = 100.0 + f64::from(i);
/// let candle = Candle::new(p, p + 1.0, p - 1.0, p, 10.0, i64::from(i)).unwrap();
/// last = evwma.update(candle);
/// }
/// assert!(last.is_some());
/// ```
#[derive(Debug, Clone)]
pub struct Evwma {
period: usize,
/// Rolling window of `(close, volume)` pairs, oldest at the front.
window: VecDeque<(f64, f64)>,
sum_v: f64,
current: Option<f64>,
}
impl Evwma {
/// # Errors
/// Returns [`Error::PeriodZero`] if `period == 0`.
pub fn new(period: usize) -> Result<Self> {
if period == 0 {
return Err(Error::PeriodZero);
}
Ok(Self {
period,
window: VecDeque::with_capacity(period),
sum_v: 0.0,
current: None,
})
}
/// Configured period.
pub const fn period(&self) -> usize {
self.period
}
/// Current value if available.
pub const fn value(&self) -> Option<f64> {
self.current
}
}
impl Indicator for Evwma {
type Input = Candle;
type Output = f64;
fn update(&mut self, candle: Candle) -> Option<f64> {
let close = candle.close;
let volume = candle.volume;
if self.window.len() == self.period {
let (_, old_v) = self.window.pop_front().expect("window is non-empty");
self.sum_v -= old_v;
}
self.window.push_back((close, volume));
self.sum_v += volume;
if self.window.len() < self.period {
return None;
}
// The volume sum may be zero (every bar in the window had zero
// volume); the recurrence is undefined, so seed/hold instead.
if self.sum_v <= 0.0 {
if self.current.is_none() {
self.current = Some(close);
}
return self.current;
}
let prev = self.current.unwrap_or(close);
let next = ((self.sum_v - volume) * prev + volume * close) / self.sum_v;
self.current = Some(next);
Some(next)
}
fn reset(&mut self) {
self.window.clear();
self.sum_v = 0.0;
self.current = None;
}
fn warmup_period(&self) -> usize {
self.period
}
fn is_ready(&self) -> bool {
self.current.is_some()
}
fn name(&self) -> &'static str {
"EVWMA"
}
}
#[cfg(test)]
mod tests {
use super::*;
use crate::traits::BatchExt;
use approx::assert_relative_eq;
fn candle(close: f64, volume: f64, ts: i64) -> Candle {
Candle::new(close, close, close, close, volume, ts).unwrap()
}
#[test]
fn rejects_zero_period() {
assert!(matches!(Evwma::new(0), Err(Error::PeriodZero)));
}
#[test]
fn accessors_and_metadata() {
let mut e = Evwma::new(5).unwrap();
assert_eq!(e.period(), 5);
assert_eq!(e.warmup_period(), 5);
assert_eq!(e.name(), "EVWMA");
assert_eq!(e.value(), None);
for i in 0..5 {
e.update(candle(10.0, 1.0, i));
}
assert!(e.value().is_some());
}
#[test]
fn constant_series_yields_the_constant() {
// A flat close — every (V_sum - v) * prev + v * close reduces to
// V_sum * close, so the recurrence preserves the constant after the
// first seeded sample.
let mut e = Evwma::new(5).unwrap();
let candles: Vec<Candle> = (0..30).map(|i| candle(42.0, 3.0, i)).collect();
let out = e.batch(&candles);
for v in out.iter().skip(4).flatten() {
assert_relative_eq!(*v, 42.0, epsilon = 1e-12);
}
}
#[test]
fn reference_value_period_2() {
// EVWMA(2). Bars: (close, volume) = (10, 1), (20, 3), (30, 1).
// Bar 1: window not full (size 1) -> None.
// Bar 2: window full, sum_v = 4, prev seeds to 20.
// EVWMA = ((4 - 3) * 20 + 3 * 20) / 4 = 80 / 4 = 20.
// Bar 3: window slides, sum_v = 4 (drops the 1, gains the 1).
// EVWMA = ((4 - 1) * 20 + 1 * 30) / 4 = (60 + 30) / 4 = 22.5.
let mut e = Evwma::new(2).unwrap();
assert_eq!(e.update(candle(10.0, 1.0, 0)), None);
assert_relative_eq!(
e.update(candle(20.0, 3.0, 1)).unwrap(),
20.0,
epsilon = 1e-12
);
assert_relative_eq!(
e.update(candle(30.0, 1.0, 2)).unwrap(),
22.5,
epsilon = 1e-12
);
}
#[test]
fn warmup_emits_first_value_at_period() {
let mut e = Evwma::new(4).unwrap();
for i in 0..3 {
assert_eq!(e.update(candle(10.0, 1.0, i)), None);
}
assert!(e.update(candle(10.0, 1.0, 3)).is_some());
}
#[test]
fn zero_volume_window_holds_value() {
// Every bar has zero volume: no participation, so the recurrence
// can't move and EVWMA simply seeds to the first close.
let mut e = Evwma::new(3).unwrap();
e.update(candle(10.0, 0.0, 0));
e.update(candle(15.0, 0.0, 1));
let v = e.update(candle(20.0, 0.0, 2)).unwrap();
assert_relative_eq!(v, 20.0, epsilon = 1e-12);
// Next bar still flat-zero volume: holds 20.
let v2 = e.update(candle(50.0, 0.0, 3)).unwrap();
assert_relative_eq!(v2, 20.0, epsilon = 1e-12);
}
#[test]
fn batch_equals_streaming() {
let candles: Vec<Candle> = (0..60_i64)
.map(|i| {
let c = 100.0 + (i as f64 * 0.3).sin() * 8.0;
candle(c, 1.0 + (i % 7) as f64, i)
})
.collect();
let batch = Evwma::new(10).unwrap().batch(&candles);
let mut b = Evwma::new(10).unwrap();
let streamed: Vec<_> = candles.iter().map(|c| b.update(*c)).collect();
assert_eq!(batch, streamed);
}
#[test]
fn reset_clears_state() {
let mut e = Evwma::new(3).unwrap();
let candles: Vec<Candle> = (0..10).map(|i| candle(10.0 + i as f64, 2.0, i)).collect();
e.batch(&candles);
assert!(e.is_ready());
e.reset();
assert!(!e.is_ready());
assert_eq!(e.update(candle(10.0, 1.0, 0)), None);
}
}