Python - Behavior of `statistics.stdev()` on Numpy arrays

Question

I have a Numpy array and I would like to use statistics.stdev() to compute its standard deviation. The output of this operation is somewhat unexpected.

import numpy as np
from statistics import stdev, pstdev

>>> a = np.array([0, 1, 2, 3])
>>> stdev(a)
1.0

This is neither the sample standard deviation

>>> stdev([0, 1, 2, 3])
1.2909944487358056

nor the population standard deviation

>>> pstdev([0, 1, 2, 3])
1.118033988749895

I wonder if statistics.stdev() is not supposed to work with Numpy arrays or there is some underlying mechanism causing this behavior?

Note that np.std() works both on lists and Numpy arrays.

>>> np.std([0, 1, 2, 3])
1.118033988749895
>>> np.std(np.array([0, 1, 2, 3]))
1.118033988749895

Both give the population standard deviation as expected.

Mahdi F. · Accepted Answer · 2021-11-06 13:18:56Z

1

This happens because the type of numbers, type of numbers in the list is float64 but the type of numbers in numpy.array is int64 you can get what you get with the list with change dtype in NumPy.array to float64 like below:

>>> a = np.array([0, 1, 2, 3])
>>> a.dtype
dtype('int64')
>>> statistics.stdev(a)
1.0

>>> a = np.array([0, 1, 2, 3], dtype=float)
>>> a.dtype
dtype('float64')
>>> statistics.stdev(a)
1.2909944487358056

>>> statistics.stdev([0, 1, 2, 3])
1.2909944487358056

edited Nov 6, 2021 at 13:18

answered Nov 6, 2021 at 13:06

Mahdi F.

24.1k5 gold badges25 silver badges32 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Python - Behavior of `statistics.stdev()` on Numpy arrays

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related