TdMovingAverage

Tom DeMark's TD Moving Averages — a fast (ST1) / slow (ST2) ribbon on the median price; their relationship defines the trend.

Quick reference

Field	Value
Family	DeMark
Input type	Candle (high / low)
Output type	TdMovingAverageOutput { st1, st2 }
Output range	price units
Default parameters	(period_st1 = 5, period_st2 = 13)
Warmup period	period_st2
Interpretation	st1 > st2 uptrend; st1 < st2 downtrend; cross = change.

Formula

price = (high + low) / 2          (median price)
st1   = SMA(price, period_st1)    (fast)
st2   = SMA(price, period_st2)    (slow)

DeMark's two-line ribbon frames the trend: st1 above st2 is up, below is down, and the cross marks the change. Using the median price de-emphasises closing noise. Source: crates/wickra-core/src/indicators/td_moving_average.rs.

Parameters

Name	Type	Default	Valid range	Source	Description
period_st1	usize	5	>= 1, < period_st2	td_moving_average.rs:61	Fast average length.
period_st2	usize	13	> period_st1	td_moving_average.rs:61	Slow average length. 0 for either errors with Error::PeriodZero; bad ordering with Error::InvalidPeriod.

periods() returns (period_st1, period_st2); value returns the current output if ready.

Inputs / Outputs

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

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

A Candle in, an Option<TdMovingAverageOutput> out. The Python binding returns a (st1, st2) tuple from update and an (n, 2) array from batch(high, low); Node returns { st1, st2 } and a flat Float64Array of length n*2; WASM mirrors the object with camelCase keys.

Warmup

warmup_period() == period_st2. The slow average seeds last (first_emission_at_warmup_period pins this).

Edge cases

• Uptrend. Fast leads slow (fast_leads_slow_in_uptrend pins this).
• Downtrend. Fast trails slow (fast_below_slow_in_downtrend pins this).
• Flat series. Both lines equal the price (flat_series_equal_lines pins this).
• Finiteness. Candle::new rejects non-finite fields, so no in-method guard is needed.
• Reset. td.reset() clears both averages and the last value (reset_clears_state).

Examples

Rust

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

fn main() -> Result<(), Box<dyn std::error::Error>> {
    let mut td = TdMovingAverage::new(3, 7)?;
    let candles: Vec<Candle> = (0..40)
        .map(|i| { let b = 100.0 + f64::from(i); Candle::new(b, b + 1.0, b - 1.0, b, 1_000.0, 0).unwrap() })
        .collect();
    let out = td.batch(&candles).last().unwrap().unwrap();
    println!("st1 > st2: {}", out.st1 > out.st2);
    Ok(())
}

Output:

st1 > st2: true

Python

import numpy as np
import wickra as ta

td = ta.TdMovingAverage(5, 13)
high = np.arange(40, dtype=float) + 101
low  = np.arange(40, dtype=float) + 99
st1, st2 = td.batch(high, low).T
print(st1[-1], st2[-1])

Node

const ta = require('wickra');

const td = new ta.TdMovingAverage(5, 13);
console.log('warmupPeriod:', td.warmupPeriod()); // 13

Streaming

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

let mut td = TdMovingAverage::new(5, 13).unwrap();
let feed: Vec<Candle> = Vec::new(); // your live stream
for candle in feed {
    if let Some(out) = td.update(candle) {
        // out.st1 crossing out.st2 -> trend change
    }
}

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

Interpretation

1. Trend ribbon. Trade with the ribbon: longs while st1 > st2, shorts while below.
2. Crossover timing. The ST1/ST2 cross is the DeMark trend-change trigger.
3. Median basis. Because it uses the median price, whipsaws from erratic closes are reduced versus a close-based MA cross.

Common pitfalls

• Lagging cross. Like any MA cross, it confirms after the turn; combine with a faster signal for entries.
• Choppy markets. The cross whipsaws in ranges; gate with a trend/volatility filter.
• Median, not close. Signals will not match a close-based dual SMA exactly.

References

DeMark, T. R. (1994), The New Science of Technical Analysis; Perl, J. (2008), DeMark Indicators, Bloomberg Press.

See also

• Indicator-Sma — the simple moving average building block.
• Indicator-MedianPrice — the (high+low)/2 input.
• Indicator-MacdIndicator — momentum of an MA difference.
• Indicators-Overview — the full taxonomy.