RTTADocumentation Built by Goblin Reactor ↗

TrendIntensityIndex

Incremental, causal technical analysis documentation

Summary

TrendIntensityIndex is RTTA's streaming Trend Intensity Index (TII): the percentage of absolute close-to-SMA deviations that are positive over a rolling window, scaled to \(0\)–\(100\).

Update API

value = rtta.TrendIntensityIndex(window=30, fillna=True).update(close)

With fillna=False, output is NaN until window samples have been seen.

Theory Of Operation

TII measures how consistently price sits above its moving average. For each bar the signed residual \(close - SMA\) is split into a positive part and its absolute value; rolling sums of those quantities form a ratio. A reading near 100 means price has been almost exclusively above the SMA (strong uptrend intensity); near 0 means almost exclusively below; near 50 means a balanced mix.

Recurrence

Let \(c_t\) be close and \(n\) be window (default \(30\)).

\[S_t = \operatorname{SMA}_n(c_t), \qquad d_t = c_t - S_t\]
\[p_t = \max(d_t, 0), \qquad a_t = |d_t|\]

Maintain rolling sums over the last \(\min(t, n)\) samples:

\[P_t = \sum_{i \in W_t} p_i, \qquad A_t = \sum_{i \in W_t} a_i\]
\[TII_t = 100 \cdot \frac{P_t}{A_t} \quad\text{(safe divide)}\]

The nested SMA is constructed with fillna=True. Outer warm count is \(n\).

Implementation Notes

The recurrence is implemented in src/rtta/indicator.cpp in class TrendIntensityIndex with members sma_, pos_, and abs_.

Reference