Description
Code Sample, a copy-pastable example if possible
import pandas as pd
import numpy as np
a = pd.DataFrame([[1, 2, 3], [4, 5, 6], [7, 8, 9]]).set_index([0, 1])
b = pd.DataFrame([[1, np.nan, 3], [4, np.nan, 6], [7, np.nan, 9]]).set_index([0, 1])
a_idx = a.index[1]
b_idx = b.index[1]>>> a.loc[a_idx:]
2
0 1
4 5 6
7 8 9
>>> a.loc[:a_idx]
2
0 1
1 2 3
4 5 6
>>> b.loc[b_idx:]
2
0 1
7 NaN 9
>>> b.loc[:b_idx]
2
0 1
1 NaN 3
4 NaN 6
Problem description
As above, slicing on MultiIndex has different behaviours with nans: it excludes the current index of selection from the slice. This is possibly because - rightly so - there is no match on NaNs when comparing the indices, but if that is the case, this should also not work:
>>> b.loc[b_idx]
2 6
Name: (4, nan), dtype: int64
Note: might be a duplicate of #25154, but OP does not slice directly on the MultiIndex with NaNs, so I am not sure if fixing that would also fix this issue.
Expected Output
>>> b.loc[b_idx:]
2
0 1
4 NaN 6
7 NaN 9
Output of pd.show_versions()
INSTALLED VERSIONS
commit : None
python : 3.7.4.final.0
python-bits : 64
OS : Windows
OS-release : 10
machine : AMD64
processor : Intel64 Family 6 Model 158 Stepping 10, GenuineIntel
byteorder : little
LC_ALL : None
LANG : None
LOCALE : None.None
pandas : 0.25.1
numpy : 1.16.4
pytz : 2019.2
dateutil : 2.8.0
pip : 19.2.2
setuptools : 41.0.1
Cython : 0.29.13
pytest : 5.1.2
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : 1.1.8
lxml.etree : None
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : 2.10.1
IPython : 7.8.0
pandas_datareader: None
bs4 : None
bottleneck : 1.2.1
fastparquet : None
gcsfs : None
lxml.etree : None
matplotlib : 3.1.1
numexpr : None
odfpy : None
openpyxl : 2.6.2
pandas_gbq : None
pyarrow : None
pytables : None
s3fs : None
scipy : 1.3.1
sqlalchemy : 1.3.7
tables : None
xarray : 0.12.3
xlrd : 1.2.0
xlwt : None
xlsxwriter : 1.1.8