QuoteStuffingDetector

Summary

QuoteStuffingDetector is RTTA's streaming implementation of: EWMA quote-to-trade message ratio detector for quote-stuffing episodes.

Update API

result = rtta.QuoteStuffingDetector().update(quote_messages, trades)

The update(...) call consumes one observation using quote_messages, trades. advance(...) uses the same inputs when the caller wants to update state without materializing a Python return value.

Theory Of Operation

QuoteStuffingDetector first constructs a scalar market-state metric from the current observation and compact streaming state, then passes that metric through explicit entry/exit hysteresis. The metric is named in the recurrence below; the hysteresis keeps the output stable until the metric crosses the opposite exit band.

Recurrence

Let \(z_t = (quote_messages_t, trades_t)\) denote the observation consumed by one update(...) call and let \(\theta\) denote constructor parameters such as window lengths, thresholds, and smoothing constants.

\[\rho_t=\frac{\max(quote\_messages_t,0)}{\max(\max(trades_t,0),\epsilon)}\]

\[q_t=\alpha\rho_t+(1-\alpha)q_{t-1}\]

\[r_t = \begin{cases} 1, & r_{t-1} = 0 \text{ and } q_t \ge e \\ 0, & r_{t-1} = 1 \text{ and } q_t \le x \\ r_{t-1}, & \text{otherwise} \end{cases}, \qquad x < e\]

The return value is the current scalar indicator value.

Implementation Notes

The recurrence is implemented in src/rtta/indicator.cpp in class QuoteStuffingDetector.

Reference

Background reference