KST

Martin Pring's Know Sure Thing — a long-horizon momentum oscillator combining four SMA-smoothed Roc series with Pring's fixed weights 1, 2, 3, 4, plus an SMA signal line.

Quick reference

Field	Value
Family	Momentum Oscillators
Input type	f64 (single close)
Output type	KstOutput { kst, signal }
Output range	unbounded around zero
Default parameters	Pring's classic (10, 15, 20, 30, 10, 10, 10, 15, 9)
Warmup period	max(roc_i + sma_i) + signal_period − 1 (53 for classic)
Interpretation	A cross of the kst line above its signal is bullish; below is bearish.

Formula

RCMA_i = SMA(ROC(close, roc_i), sma_i)        for i = 1..=4
KST    = 1·RCMA_1 + 2·RCMA_2 + 3·RCMA_3 + 4·RCMA_4
Signal = SMA(KST, signal_period)

Kst::classic() applies Pring's recommended defaults: ROC lookbacks (10, 15, 20, 30), RCMA smoothing (10, 10, 10, 15), signal 9. The ascending 1·…4· weights make the longest-lookback ROC dominate, so KST is a long-term momentum gauge — slower and smoother than a single ROC or MACD.

Parameters

Nine usize parameters, all required non-zero (zero returns [Error::PeriodZero]):

Name	Default	Description
roc1 … roc4	10, 15, 20, 30	Lookback for the four ROC series.
sma1 … sma4	10, 10, 10, 15	Smoothing length for each RCMA_i.
signal	9	Smoothing length for the signal line.

Source: Kst::new (kst.rs:72). Kst::classic() returns (10, 15, 20, 30, 10, 10, 10, 15, 9). Python defaults come from the pyo3 signature; the Node constructor takes all nine explicitly. The public class is KST in both bindings.

Inputs / Outputs

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

• Python. update(value) returns (kst, signal) or None; batch(prices) returns an (n, 2) np.ndarray with columns [kst, signal]; warmup rows are NaN.
• Node. update(value) returns a { kst, signal } object or null; batch(prices) returns a flat Array<number> of length 2n, interleaved [kst0, signal0, …].

Warmup

warmup_period() returns max_i(roc_i + sma_i) + signal_period − 1. Each RCMA branch is ready once its inner ROC has warmed (roc_i + 1 inputs) and its SMA has filled; the four run in parallel, so the slowest branch dominates, and the signal SMA adds signal_period − 1 on top. For the classic config the slowest branch is ROC(30) + SMA(15) = 45, plus 9 − 1 = 8, so KST emits at input 53. Pinned by warmup_emits_first_value_at_warmup_period.

Edge cases

• Constant series. ROC is 0 on a flat series, so every RCMA and the KST itself are 0, and the signal inherits 0 (test constant_series_yields_zero).
• Pure uptrend. Every ROC > 0 ⇒ every RCMA > 0 ⇒ KST and signal > 0 (test pure_uptrend_is_positive).
• Reset. reset() resets all four ROCs, all five SMAs and both held values.

Examples

Rust

use wickra::{BatchExt, Indicator, Kst};

fn main() -> Result<(), Box<dyn std::error::Error>> {
    let prices: Vec<f64> = (1..=120).map(|i| f64::from(i) * 2.0).collect();
    let mut kst = Kst::classic();
    if let Some(o) = kst.batch(&prices).into_iter().flatten().last() {
        println!("kst={:.4} signal={:.4}", o.kst, o.signal);
    }
    Ok(())
}

Python

import numpy as np
import wickra as ta

kst = ta.KST(10, 15, 20, 30, 10, 10, 10, 15, 9)
out = kst.batch(prices)  # shape (n, 2): [kst, signal]
line, signal = out[:, 0], out[:, 1]

Node

const ta = require('wickra');
const kst = new ta.KST(10, 15, 20, 30, 10, 10, 10, 15, 9);
const o = kst.update(101.5); // null during warmup, else { kst, signal }

Interpretation

KST reads long-term momentum by stacking four rate-of-change windows:

1. Signal-line crossovers. The primary signal is the kst line crossing its signal SMA — up is bullish, down is bearish. Because of the long lookbacks these are infrequent, high-conviction turns.
2. Zero-line and divergence. KST above zero confirms an uptrend regime; divergence between price and KST flags weakening momentum before price turns.

Common pitfalls

• Using it for short-term timing. KST is deliberately slow (classic warmup 53 bars); it is a regime/position filter, not a scalping trigger.
• Tuning all nine knobs blindly. Pring's weights and periods are balanced; change them only with a clear reason, or stick to classic().

References

• Martin J. Pring, "Summed Rate of Change (KST)", Technical Analysis of Stocks & Commodities, 1992.

See also

• Roc — the building block KST sums.
• Coppock — another weighted-ROC long-term momentum curve.
• Pmo — the DecisionPoint Price Momentum Oscillator, a related smoothed-ROC gauge.