rolling.sum() of zeros produce positive output w/ float64

[someben]

Code Sample, a copy-pastable example if possible

import io

import numpy as np
import pandas as pd

def test_rolling_sum(float_type):
    err_df = pd.DataFrame([[
        123.0,
        0.0,
        1.123456789,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
    ]], dtype=float_type).T
    assert err_df.rolling(7).sum().iloc[-1, 0] == 0.0

print(pd.__version__)
test_rolling_sum(np.float32)
test_rolling_sum(np.float64)

Problem description

The sum of a window of float32 zeros is zero, but the sum of a window of float64 zeros is a tiny, positive number. This is concerning, because ratio calculations like growth rates can "blow up" with tiny numbers.

Expected Output

0.23.4

Output of pd.show_versions()

INSTALLED VERSIONS

commit: None
python: 3.6.6.final.0
python-bits: 64
OS: Linux
OS-release: 3.13.0-103-generic
machine: x86_64
processor: x86_64
byteorder: little
LC_ALL: None
LANG: en_US.UTF-8
LOCALE: en_US.UTF-8

pandas: 0.23.4
pytest: 3.9.1
pip: 18.1
setuptools: 39.1.0
Cython: None
numpy: 1.15.3
scipy: 1.1.0
pyarrow: 0.11.0
xarray: 0.10.9
IPython: 7.0.1
sphinx: None
patsy: 0.5.1
dateutil: 2.7.3
pytz: 2018.5
blosc: None
bottleneck: 1.2.1
tables: 3.4.4
numexpr: 2.6.8
feather: None
matplotlib: 3.0.1
openpyxl: None
xlrd: None
xlwt: None
xlsxwriter: None
lxml: None
bs4: 4.6.3
html5lib: 1.0.1
sqlalchemy: None
pymysql: None
psycopg2: 2.7.5 (dt dec pq3 ext lo64)
jinja2: 2.10
s3fs: None
fastparquet: None
pandas_gbq: None
pandas_datareader: None

[chris-b1]

Thanks for the report - this is the same issue as #13254 - would definitely take a PR to fix!