numpy array elementwise multiply a panda timeseries

Question

I have these two data structures:

a = np.array([1,2,3])
ts = pd.TimeSeries([1,2,3])

What I want to get at the end is:

1  2  3
2  4  6
3  6  9

askewchan · Accepted Answer · 2014-03-12 16:45:03Z

5

You can use the outer product:

In [490]: np.outer(a, ts)
Out[490]: 
array([[1, 2, 3],
       [2, 4, 6],
       [3, 6, 9]])

Or align one of them vertically first:

In [491]: a * ts[:, None]
Out[491]: 
array([[1, 2, 3],
       [2, 4, 6],
       [3, 6, 9]])

Note that the strange index just makes it a column vector:

In [493]: ts[:, None]
Out[493]: 
array([[1],
       [2],
       [3]])

by adding an extra length-1 dimension to the shape:

In [494]: ts[:, None].shape
Out[494]: (3, 1)

answered Mar 12, 2014 at 16:45

askewchan

46.7k18 gold badges125 silver badges135 bronze badges

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

Comments

Jake Burkhead · Accepted Answer · 2014-03-12 16:44:36Z

4

>>> np.outer(a, ts)
array([[1, 2, 3],
       [2, 4, 6],
       [3, 6, 9]])

answered Mar 12, 2014 at 16:44

Jake Burkhead

6,5452 gold badges24 silver badges32 bronze badges