The predictive power of these features in ML models is comparable to FMA which was 0.529 and to EMA which is 0.526. These do not offer any upside in predictability of returns

var Laguerre(var *Data, var alpha)
{
var *L = series(Data[0]);
L[0] = alpha*Data[0] + (1-alpha)*L[1];
L[2] = -(1-alpha)*L[0] + L[1] + (1-alpha)*L[2+1];
L[4] = -(1-alpha)*L[2] + L[2+1] + (1-alpha)*L[4+1];
L[6] = -(1-alpha)*L[4] + L[4+1] + (1-alpha)*L[6+1];
return (L[0]+2*L[2]+2*L[4]+L[6])/6;
}

Can you also please share the logic for BandPass filter similarly?

However there are some small definition typos in your post.
In the paragraph ‘Comparing Impulse Responses’ it would be advisable to rename the text ‘impulse response’ to ‘unit step response’ or more simply ‘step response’.

An ‘impulse response’ is a reaction to a spike (impulse) of a certain amplitude.
An ‘unit impulse response’ is a reaction to a spike of amplitude 1.0.
A ‘step response’ is a reaction to a step,
An ‘unit step response’ is a reaction to a step of amplitude 1.0