ParticleFilterTrend

Incremental, causal technical analysis documentation

Summary

ParticleFilterTrend is RTTA's streaming implementation of: Deterministic-seed particle trend filter with Laplace measurement likelihood and effective sample size output.

Update API

result = rtta.ParticleFilterTrend(particles=64).update(close)

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

Theory Of Operation

ParticleFilterTrend implements the streaming form of Deterministic-seed particle trend filter with Laplace measurement likelihood and effective sample size output. Each update(...) call consumes exactly one new observation tuple and advances the internal state before returning the current value or result struct.

Recurrence

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

\[x_t^{(i)} = f(x_{t-1}^{(i)})+\epsilon_t^{(i)}\]
\[w_t^{(i)} \propto w_{t-1}^{(i)}p(z_t\mid x_t^{(i)}), \qquad \sum_i w_t^{(i)}=1\]

update(...) returns a result struct with fields trend, velocity, signal, effective_sample_size.

Implementation Notes

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

Reference