GaussianFilter
Multi-pole Gaussian low-pass filter for reduced lag.
Usage
Use when smooth symmetric price averaging with near-zero phase shift is needed. Works well as a preprocessing step for spectral analysis indicators.
Background
Gaussian filters are the theoretically optimal lowpass filter for minimizing the product of time-domain duration and frequency-domain bandwidth. Ehlers implements them as cascaded pole filters with Gaussian-function-derived coefficients, achieving very smooth output with excellent stopband attenuation.
Parameters
period(default: 14): Critical periodpoles(default: 4): Number of poles (1-4)
Formula
[ \alpha = -\beta + \sqrt{\beta^2 + 2\beta} ] [ \beta = \frac{1 - \cos(2\pi/P)}{2^{1/(2N)} - 1} ]