TII

M.H. Pee's Trend Intensity Index — the share of the most recent dev_period SMA-deviations that sit on the positive side, scaled to [0, 100]. Saturates at the extremes on sustained trends.

Quick reference

Field	Value
Family	Trend & Directional
Input type	f64 (single close)
Output type	f64
Output range	[0, 100]
Default parameters	sma_period = 60, dev_period = 30
Warmup period	sma_period + dev_period − 1 (89 for defaults)
Interpretation	> 80 strong uptrend, < 20 strong downtrend, 50 flat / no trend.

Formula

Given an SMA(close, sma_period), for each bar t past warmup compute the deviation dev_t = close_t − SMA_t. Then over the most recent dev_period deviations:

SD_pos = Σ_{i in window, dev_i > 0}  dev_i
SD_neg = Σ_{i in window, dev_i < 0}  |dev_i|
TII    = 100 · SD_pos / (SD_pos + SD_neg)

A perfectly flat window (every dev_i == 0) returns the neutral mid-point 50. The output is clamped to [0, 100] against floating-point drift in the rolling subtractions. Pee's canonical setup uses dev_period = sma_period / 2.

Parameters

Name	Type	Default	Constraint	Source
sma_period	usize	60	>= 1	Tii::new (tii.rs:66)
dev_period	usize	30	>= 1	tii.rs:66

Either parameter == 0 returns [Error::PeriodZero]. Python defaults come from #[pyo3(signature = (sma_period=60, dev_period=30))]; the Node constructor takes both explicitly. The public class is TII in both bindings.

Inputs / Outputs

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

A single f64 close in, an Option<f64> out in [0, 100]. Python maps this to float | None / a float64 np.ndarray with NaN warmup; Node to number | null / Array<number>.

Warmup

warmup_period() returns sma_period + dev_period − 1. The SMA emits at input sma_period; the deviation ring then needs dev_period − 1 more inputs to fill, so the first TII lands at sma_period + dev_period − 1 (89 for the defaults). Pinned by first_emission_at_warmup_period.

Edge cases

• Pure uptrend. Every close sits above the lagging SMA ⇒ every deviation positive ⇒ TII = 100 (test pure_uptrend_saturates_at_100).
• Pure downtrend. Every close below the SMA ⇒ TII = 0 (test pure_downtrend_falls_to_zero).
• Flat series. Every deviation 0 ⇒ neutral 50 (test constant_series_yields_neutral_50).
• Bounded. Always within [0, 100] (test output_bounded_in_unit_interval).
• Reset. reset() clears the SMA, the deviation window and both sums.

Examples

Rust

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

fn main() -> Result<(), Box<dyn std::error::Error>> {
    // Strict uptrend: every close above the lagging SMA → TII saturates at 100.
    let prices: Vec<f64> = (1..=80).map(|i| 100.0 + f64::from(i)).collect();
    let mut t = Tii::new(10, 5)?;
    println!("{:?}", t.batch(&prices).into_iter().flatten().last()); // Some(100.0)
    Ok(())
}

Python

import numpy as np
import wickra as ta

t = ta.TII(60, 30)
out = t.batch(np.array([...], dtype=float))  # 1-D in [0, 100], NaN for first 88 rows

Node

const ta = require('wickra');
const t = new ta.TII(60, 30);
const v = t.update(101.5); // null during warmup, else a value in [0, 100]

Interpretation

TII frames "how one-sided is price relative to its trend?" as a 0–100 percentage:

1. Trend confirmation. Above 80 the recent closes are overwhelmingly above the SMA — a confirmed uptrend; below 20 a confirmed downtrend.
2. Range detection. Readings hovering near 50 mean positive and negative deviations roughly cancel — a trendless, mean-reverting regime where trend-following entries should be avoided.

Unlike Adx, TII is referenced to a moving average, so it answers a slightly different question: not "is there a trend?" but "is price persistently on one side of its mean?".

Common pitfalls

• Reading direction below the SMA. A low TII means closes are mostly below the SMA — it is directional, unlike ADX which is direction-agnostic.
• Using a tiny sma_period. Pee's design is long-horizon (60/30); short periods make the deviation sign flip constantly and the index hover near 50.

References

• M.H. Pee, "Trend Intensity Index", Technical Analysis of Stocks & Commodities, June 2002.

See also

• Adx — directional-strength complement (no MA reference).
• VerticalHorizontalFilter — range-based trend-strength filter.
• ChoppinessIndex — trend-vs-range gauge.