RTTADocumentation Built by Goblin Reactor ↗

EDDM

Incremental, causal technical analysis documentation

Summary

EDDM is RTTA's streaming implementation of: Early Drift Detection Method using distances between prediction errors.

Update API

result = rtta.EDDM().update(error)

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

Theory Of Operation

EDDM is a streaming classifier-error drift detector. It treats positive input values as errors and compares the current error process against the best historical baseline using the detector's bound: binomial standard error for DDM, distance-between-errors degradation for EDDM, and a Hoeffding bound for HDDM.

Recurrence

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

\[d_k=i_k-i_{k-1} \quad \text{for the sample indices } i_k \text{ where } error_{i_k}>0\]
\[\bar{d}_k=\bar{d}_{k-1}+\frac{d_k-\bar{d}_{k-1}}{k}, \qquad s^2_{d,k}=\frac{1}{k-1}\sum_{j=1}^{k}(d_j-\bar{d}_k)^2\]
\[M_k=\bar{d}_k+2s_{d,k}, \qquad \rho_k=\frac{M_k}{\max_{j\le k}M_j}\]
\[y_t = \begin{cases} 1, & \rho_k < \rho_{drift}\\ 0.5, & \rho_k < \rho_{warning}\\ 0, & \text{otherwise} \end{cases}\]

The return value is the current scalar indicator value.

Implementation Notes

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

Reference