Time Segmented Volume (TSV)

Don Worden's Time Segmented Volume — a rolling sum of signed volume weighted by the bar's close-to-close move. Summed over a fixed window, the result quantifies the net accumulation (positive) or distribution (negative) over that span.

Quick reference

Item	Value
Family	Volume
Input type	Candle (uses close, volume)
Output type	f64
Output range	unbounded; centred near zero
Default parameters	period required (Worden's default 18)
Warmup period	period + 1
Interpretation	Net accumulation/distribution over the window

Formula

flow_t = (close_t - close_{t-1}) · volume_t     (signed money flow)
TSV_t  = Σ_{i=t-period+1..t} flow_i              (rolling window sum)

See crates/wickra-core/src/indicators/tsv.rs.

Parameters

Name	Type	Default	Constraint	Description
period	usize	none	> 0	Rolling sum window. Worden's typical 18.

Inputs / Outputs

Indicator<Input = Candle, Output = f64>. Python: Tsv(period).batch(close, volume) returns a 1-D np.ndarray with NaN warmup. Node: same shape.

Warmup

warmup_period() == period + 1. Needs period flow values, which need period + 1 prices.

Edge cases

• Flat close. Zero flow contribution.
• High-volume distribution. Single down-bar on huge volume drives TSV sharply negative.
• Reset. Clears the rolling window and prior close.

Examples

Rust

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

fn main() -> Result<(), Box<dyn std::error::Error>> {
    let candles: Vec<Candle> = (0..40).map(|i| {
        let b = 100.0 + (f64::from(i) * 0.3).sin() * 5.0;
        Candle::new(b, b + 0.5, b - 0.5, b, 1000.0, i as i64).unwrap()
    }).collect();
    let mut t = Tsv::new(18)?;
    println!("row 30 = {:?}", t.batch(&candles)[30]);
    Ok(())
}

Python

import numpy as np
import wickra as ta

n = 40
close = 100 + np.sin(np.linspace(0, 12, n)) * 5
vol = np.full(n, 1000.0)
t = ta.Tsv(18)
print(t.batch(close, vol)[30])

Node

const wickra = require('wickra');
const t = new wickra.Tsv(18);
// feed c, v

Streaming

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

let mut t = Tsv::new(18).unwrap();
let candle_stream: Vec<wickra::Candle> = Vec::new(); // your live OHLCV candle feed
for bar in candle_stream {
    if let Some(v) = t.update(bar) {
        if v > 0.0 { /* net accumulation over period */ }
        if v < 0.0 { /* net distribution */ }
    }
}

Interpretation

• Sign of TSV. Positive = net accumulation over the period, negative = net distribution. Crosses of zero signal accumulation/distribution regime changes.
• Vs OBV. OBV uses just ±volume (sign by direction only); TSV multiplies by close-change magnitude — gives weight to large moves.
• Worden pairing. Worden's published rules pair TSV with a 9-bar EMA of itself as the signal line; crossover triggers buy / sell signals.

Common pitfalls

• Comparing magnitudes across instruments. TSV scales with volume; tickers with very different volume profiles produce incomparable TSV magnitudes. Use slope / direction, not absolute value.
• Period choice. 18 is Worden's default; shorter periods (10-12) make TSV more responsive but noisier.

References

• Don Worden (Worden Brothers), 1980s — original TSV formulation for the TC2000 charting package.

See also

• Obv — simpler ±volume cumulative.
• Kvo — alternative volume-momentum oscillator.
• Adl — Chaikin's accumulation/distribution.
• Indicators-Overview — full taxonomy.