VolumeWeightedSr

A support/resistance band whose edges are the volume-weighted average of recent lows and highs — levels gravitate to where trading actually happened.

Quick reference

Field	Value
Family	Pivots & S/R
Input type	Candle (high / low / volume)
Output type	VolumeWeightedSrOutput { support, resistance }
Output range	price units; support <= resistance
Default parameters	(period = 20) (Python)
Warmup period	period
Interpretation	Volume-weighted band; edges are sticky S/R levels.

Formula

support    = Σ(low_i  · volume_i) / Σ volume_i      over the window
resistance = Σ(high_i · volume_i) / Σ volume_i      over the window

Plain high/low channels weight every bar equally, so a thin wick sets the boundary. Volume-weighting pulls the support and resistance toward the lows and highs that carried real volume — the prices the market agreed mattered — producing levels that tend to hold better. If the window's total volume is zero the band falls back to the equal-weighted average high and low. Source: crates/wickra-core/src/indicators/volume_weighted_sr.rs.

Parameters

Name	Type	Default	Valid range	Source	Description
period	usize	20 (Python)	>= 1	volume_weighted_sr.rs:75	Lookback window. 0 errors with Error::PeriodZero.

The period getter returns the window; value returns the current output if ready.

Inputs / Outputs

From crates/wickra-core/src/indicators/volume_weighted_sr.rs:

use wickra::{Candle, Indicator, VolumeWeightedSr, VolumeWeightedSrOutput};
// VolumeWeightedSr: Input = Candle, Output = VolumeWeightedSrOutput
const _: fn(&mut VolumeWeightedSr, Candle) -> Option<VolumeWeightedSrOutput> =
    <VolumeWeightedSr as Indicator>::update;

A Candle in, an Option<VolumeWeightedSrOutput> out. The Python binding returns a (support, resistance) tuple from update and an (n, 2) array from batch(high, low, volume); Node returns { support, resistance } and a flat Float64Array of length n*2; WASM mirrors the object with camelCase keys.

Warmup

warmup_period() == period. The first value lands once the window is full (first_emission_at_warmup_period pins this).

Edge cases

• Ordering. support <= resistance always (support_below_resistance pins this).
• Volume pull. A heavy bar drags the band toward its high/low (weights_toward_high_volume_bars pins this).
• Zero volume. A volume-free window falls back to equal-weighted means (zero_volume_falls_back_to_equal_weight pins this).
• Finiteness. Candle::new rejects non-finite fields, so no in-method guard is needed.
• Reset. v.reset() clears the windows, the running sums and the last value (reset_clears_state).

Examples

Rust

use wickra::{BatchExt, Candle, Indicator, VolumeWeightedSr};

fn main() -> Result<(), Box<dyn std::error::Error>> {
    let mut v = VolumeWeightedSr::new(4)?;
    let c = |h: f64, l: f64, vol: f64| Candle::new(l, h, l, (h+l)/2.0, vol, 0).unwrap();
    let candles = [c(102.0,98.0,100.0), c(102.0,98.0,100.0), c(102.0,98.0,100.0), c(112.0,108.0,9_000.0)];
    let out = v.batch(&candles).last().unwrap().unwrap();
    println!("resistance leans to heavy bar: {:.1}", out.resistance);
    Ok(())
}

Output:

resistance leans to heavy bar: 11x.x

Python

import numpy as np
import wickra as ta

v = ta.VolumeWeightedSr(20)
high = ...; low = ...; volume = ...
support, resistance = v.batch(high, low, volume).T

Node

const ta = require('wickra');

const v = new ta.VolumeWeightedSr(20);
console.log('warmupPeriod:', v.warmupPeriod()); // 20

Streaming

use wickra::{Candle, Indicator, VolumeWeightedSr};

let mut v = VolumeWeightedSr::new(20).unwrap();
let mut last = None;
for i in 0..40 {
    let base = 100.0 + (f64::from(i) * 0.3).sin() * 5.0;
    let c = Candle::new(base, base + 2.0, base - 2.0, base, 1_000.0 + f64::from(i), 0).unwrap();
    last = v.update(c);
}
println!("{last:?}");

Streaming update and batch are equivalent tick-for-tick (batch_equals_streaming pins this).

Interpretation

1. Sticky levels. Trade reversals off the support/resistance edges; they reflect volume-validated prices rather than wick extremes.
2. Band width. A widening band signals expanding, volume-backed range; a narrowing one signals compression.
3. Confluence. Overlap with pivots or a volume profile POC for higher-conviction levels.

Common pitfalls

• Needs real volume. On feeds with synthetic volume the weighting is meaningless; it then equals a plain average band.
• Not the extreme. These are weighted averages of highs/lows, not the max/min; for the absolute channel use Donchian.
• Window choice. Longer windows give stabler but laggier levels.

References

Volume-weighting of price levels follows the same logic as VWAP (Berkowitz, Logue & Noser, 1988); the high/low band form is a common practitioner construction.

See also

• Indicator-Donchian — equal-weighted high/low channel.
• Indicator-Vwap — volume-weighted average price.
• Indicator-VolumeProfile — full volume-at-price profile.
• Indicators-Overview — the full taxonomy.