boxema v0.0.0

Box filter based Exponential Moving Average

This EMA is designed to be identical to the EMA given in http://matplotlib.org/examples/pylab_examples/finance_work2.html however it is rolling (can be updated at any time). Because it is a rolling EMA, beware that the first n results will be different from the python example. This is because the python example uses result n + 1 for all results 0-n and since this algorithm is rolling, that is not possible. Discard the first n results and all following result will be the same.

To use:

const Ema = require('boxema');

// Create an EMA with period 8
const ema8 = Ema.create(8);

[1,3,5,7,9].forEach((x) => {
    console.log("EMA is now " + ema8.update(x));
});

console.log("it can also be used as a list converter:");
console.log([2,4,6,8].map(ema8));

console.log("you can get the last result using ema.get()");
console.log(ema8.get());

The python code on which this is based is as follows, note the commented line which is not possible with a rolling EMA.

def moving_average(x, n):
    weights = np.exp(np.linspace(-1., 0., n))
    weights /= weights.sum()
    a =  np.convolve(x, weights, mode='full')[:len(x)]
    #a[:n] = a[n]  ## This is not possible with a rolling EMA so it is removed.
    return a