Summary
PsychologicalLine is RTTA's streaming Psychological Line (PSY): the percentage
of bars in a rolling window where close rose versus the prior close.
Update API
value = rtta.PsychologicalLine(window=12, fillna=True).update(close)
The first sample has no prior close, so its up-flag is \(0\). With
fillna=False, output is NaN until the window buffer is full.
Theory Of Operation
PSY is a simple market-sentiment / breadth-style oscillator for a single price series: it counts how often the market "won" (up close) over the last \(n\) bars. Readings near 100 mean almost every bar closed higher; near 0 means almost every bar closed lower. The classic interpretation treats extreme highs as crowded bullish conditions and extreme lows as crowded bearish conditions.
Recurrence
Let \(c_t\) be close and \(n\) be window (default \(12\)). Define the up
indicator:
Maintain a rolling sum \(U_t = \sum_{i \in W_t} u_i\) over the last \(\min(t+1, n)\) flags (FIFO buffer). Let \(m_t\) be the current buffer size (at least 1 in the formula path):
Implementation Notes
The recurrence is implemented in src/rtta/indicator.cpp in
class PsychologicalLine using rolling_sum_push on buffer ups_.