StochasticCCI

Stochastic CCI — a stochastic oscillator computed over the CCI, re-scaling the unbounded CCI into a bounded [0, 100] momentum oscillator.

Quick reference

Field	Value
Family	Momentum Oscillators
Input type	Candle (typical price = (high + low + close) / 3)
Output type	f64 (%K)
Output range	[0, 100]; neutral 50 in a flat market
Default parameters	period (shared by CCI and stochastic lookback)
Warmup period (warmup_period())	2·period − 1
Interpretation	Bounded, self-normalising CCI.

Formula

cci = CCI(typical_price, period)
%K  = 100 * (cci - lowest(cci, period)) / (highest(cci, period) - lowest(cci, period))

The CCI is unbounded and clusters inside ±100, so fixed overbought/oversold lines fit it poorly. Wrapping it in a stochastic re-expresses each CCI reading as its rank within its own recent range, producing a bounded oscillator that adapts to changing CCI volatility. The same period drives both the CCI and the stochastic window. When the CCI range over the window is zero (a flat market, CCI pinned at 0), the result is the neutral 50.

Source: crates/wickra-core/src/indicators/stochastic_cci.rs.

Parameters

Name	Type	Default	Valid range	Source	Description
period	usize	none	>= 1	stochastic_cci.rs:55	CCI period and stochastic lookback. 0 errors with Error::PeriodZero.

(Python class wickra.StochasticCCI(period); Node new ta.StochasticCCI(period).)

Inputs / Outputs

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

use wickra::{Candle, Indicator, StochasticCci};
// StochasticCci: Input = Candle, Output = f64
const _: fn(&mut StochasticCci, Candle) -> Option<f64> =
    <StochasticCci as Indicator>::update;

Node: update(high, low, close) / batch(h[], l[], c[]). Python: update(candle) / batch(high, low, close) → 1-D ndarray (NaN for warmup). Node returns number | null / Array<number> with NaN.

Warmup

warmup_period() returns 2·period − 1: the CCI seeds at bar period, then period CCI values fill the stochastic window. Pinned by accessors_and_metadata (warmup_period() == 27 for StochasticCci::new(14)) and first_emission_matches_warmup_period (== 9 for period 5, first Some at index 8).

Edge cases

• Bounded. bounded_zero_to_hundred pins every emission to [0, 100].
• Flat market → 50. Constant candles pin the CCI at 0, giving a zero range and the neutral 50; pinned by flat_market_is_neutral.
• Top of range → 100. highest_cci_in_window_is_hundred checks the latest CCI being the window maximum yields %K = 100.
• Reset. reset_clears_state. Streaming/batch: batch_equals_streaming.

Examples

Rust

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

fn main() -> Result<(), Box<dyn std::error::Error>> {
    let mut sc = StochasticCci::new(4)?;
    let bars = vec![Candle::new(10.0, 10.0, 10.0, 10.0, 1.0, 0)?; 8];
    let out: Vec<Option<f64>> = sc.batch(&bars);
    println!("warmup_period = {}", sc.warmup_period());
    println!("{:?}", out);
    Ok(())
}

Output:

warmup_period = 7
[None, None, None, None, None, None, Some(50.0), Some(50.0)]

On a flat series the CCI is pinned at 0, so its window has zero range and the oscillator returns the neutral 50 from the first emission (bar 2·period − 1 = 7).

Python

import numpy as np
import wickra as ta

sc = ta.StochasticCCI(4)
flat = np.full(8, 10.0)
out = sc.batch(flat, flat, flat)   # high, low, close
print("warmup_period =", sc.warmup_period())
print(out)

Output:

warmup_period = 7
[nan nan nan nan nan nan 50. 50.]

Node

const ta = require('wickra');
const sc = new ta.StochasticCCI(4);
const flat = Array.from({ length: 8 }, () => 10);
console.log(sc.batch(flat, flat, flat)); // high, low, close
console.log('warmupPeriod:', sc.warmupPeriod());

Output:

[ NaN, NaN, NaN, NaN, NaN, NaN, 50, 50 ]
warmupPeriod: 7

Streaming

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

let mut sc = StochasticCci::new(14)?;
let mut last = None;
for i in 0..60 {
    let base = 100.0 + (f64::from(i) * 0.3).sin() * 10.0;
    last = sc.update(Candle::new(base, base + 1.0, base - 1.0, base, 1.0, i64::from(i))?);
}
println!("{last:?}");
# Ok::<(), Box<dyn std::error::Error>>(())

Interpretation

Stochastic CCI turns the CCI into a bounded, RSI-like oscillator while keeping the CCI's sensitivity to typical-price deviations. Because it self-normalises to recent range, fixed 80/20 (or 90/10) bands work consistently across regimes where the raw CCI's ±100/±200 lines would not.

Typical uses:

1. Overbought/oversold. 80/20 crosses flag stretched CCI readings.
2. Faster turns. It tends to turn ahead of a plain CCI at extremes, since it reacts to the CCI's position in its range, not its absolute level.
3. Range vs. trend. Persistent pinning at 0 or 100 signals a strong trend; oscillation through the mid-line signals a range.

Common pitfalls

• Whipsaw at the rails. Like any stochastic, it can pin at 0/100 in a strong trend; pair with a trend filter before fading extremes.
• Double-smoothing expectations. This exposes raw %K; apply your own SMA if you want a %D signal line.

References

The stochastic-of-an-oscillator construction is a common refinement (cf. StochRsi); applied here to Lambert's CCI (1980).

See also

• Indicator-Cci — the underlying oscillator.
• Indicator-StochRsi — the same idea applied to the RSI.
• Indicator-Stochastic — the price-based stochastic.
• Indicators-Overview — the full taxonomy.