Search code examples
numpyweighted-average

Why the weighted average get unintended result?

we know the weighted average formula is:

so, when I use numpy:

a = np.array([1,2,3,4])

wts = np.array([1,2,3,4])
print(np.average(a, weights=wts))

it should be:

np.sum([1*1, 2*2, 3*3, 4*4]) / 4   # 7.5

but why get 3.0?

Solution

  • According to the doc of average, the average is

    avg = sum(a * weights) / sum(weights)

    If you want to divide by the number of weights instead of the summation, you can simply do

    a = np.array([1,2,3,4])
wts = np.array([1,2,3,4])

np.dot(a,wts) / wts.shape[0]