Recursive Median Filter

Category: trend

Combines the outlier rejection of a median filter with the smoothness of a recursive low-pass filter (essentially an EMA of the median).

Formula

\begin{cases} \omega = \frac{2\pi}{LowpassLength} \\ \alpha = \frac{\cos(\omega) + \sin(\omega) - 1}{\cos(\omega)} \\ RMF_t = \alpha \cdot Median(src, MedianLength) + (1 - \alpha) \cdot RMF_{t-1} \end{cases}

Inputs

See signal primitives, usage in published strategies and more on WOBR StrategyVerse.


Open in the WOBR AI app → · WOBR.AI home