DataFrame[list[Series[datetime64ns, tz]]] drops timezone information

[Koojav]

Code Sample, a copy-pastable example if possible

s = pd.Series([pd.to_datetime('2018-10-08 13:36:45+00:00')])


Output:
0   2018-10-08 13:36:45+00:00
dtype: datetime64[ns, UTC]

pd.DataFrame([s]).min()


Output:
0   2018-10-08 13:36:45
dtype: datetime64[ns]

Problem description

When using DataFrame().min() method timezone information gets removed but it should remain untouched just like when using Series().combine(...,min). Example:

s.combine(s, min)


Output:
0   2018-10-08 13:36:45+00:00
dtype: datetime64[ns, UTC]

Expected Output

0   2018-10-08 13:36:45+00:00
dtype: datetime64[ns, UTC]

Output of pd.show_versions()

INSTALLED VERSIONS

commit : None
python : 3.6.7.final.0
python-bits : 64
OS : Linux
OS-release : 4.15.0-64-generic
machine : x86_64
processor : x86_64
byteorder : little
LC_ALL : None
LANG : en_US.UTF-8
LOCALE : en_US.UTF-8

pandas : 0.25.0
numpy : 1.17.0
pytz : 2019.2
dateutil : 2.8.0
pip : 19.2.1
setuptools : 40.6.3
Cython : 0.27.3
pytest : 3.0.7
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : None
lxml.etree : 4.2.6
html5lib : None
pymysql : None
psycopg2 : 2.7.3.2 (dt dec pq3 ext lo64)
jinja2 : 2.10
IPython : 7.2.0
pandas_datareader: None
bs4 : 4.6.3
bottleneck : None
fastparquet : None
gcsfs : None
lxml.etree : 4.2.6
matplotlib : 2.0.2
numexpr : None
odfpy : None
openpyxl : None
pandas_gbq : 0.11.0
pyarrow : 0.13.0
pytables : None
s3fs : None
scipy : 1.0.0
sqlalchemy : 1.3.7
tables : None
xarray : None
xlrd : None
xlwt : None
xlsxwriter : None

[randomstuff]

Actually, this is the DataFrame creation which is dropping timezone, not min():

import pandas as pd
s = pd.Series([pd.to_datetime('2018-10-08 13:36:45+00:00')])
print(pd.DataFrame([s])[0])
# => 0   2018-10-08 13:36:45
# => Name: 0, dtype: datetime64[ns]
print(pd.DataFrame([s])[0].dt.tz)
# => None

[randomstuff]

But this works fine:

pd.DataFrame({"x":s}).min()
# => x   2018-10-08 13:36:45+00:00
# => dtype: datetime64[ns, UTC]

[mroeschke]

This looks fixed on master and could use a test:

In [12]: print(pd.DataFrame([s])[0])
0   2018-10-08 13:36:45+00:00
Name: 0, dtype: datetime64[ns, UTC]

In [13]: print(pd.DataFrame([s])[0].dt.tz)
UTC

In [14]: pd.__version__
Out[14]: '1.1.0.dev0+1027.g767335719.dirty'

[ghost]

What timezone should be used for, let's say, mean if the series contains datetimes with different timezones?

[jreback]

you should convert to UTC if interoperating between time zones