DataFrame.groupby(level=...).rolling(n).mean()

[691175002]

Using the groupby().rolling() object seems to duplicate a level of the index.

In [9]: d.groupby(level='ticker').rolling(30).mean()
Out[9]: 
ticker  ticker  date      
BMO     BMO     2006-01-02          NaN
                2006-01-03          NaN

TD      TD      2016-09-22    57.139340
                2016-09-23    57.171864


In [10]: d.groupby(level='ticker').apply(pd.rolling_mean, 30)
Out[10]: 
ticker  date      
BMO     2006-01-02          NaN
        2006-01-03          NaN

TD      2016-09-22    57.139340
        2016-09-23    57.171864


In [11]: d.groupby(level='ticker').apply(lambda x: x.rolling(30).mean())
Out[11]: 
ticker  date      
BMO     2006-01-02          NaN
        2006-01-03          NaN

TD      2016-09-22    57.139340
        2016-09-23    57.171864

I would expect the output to be the same in all three cases.

[jreback]

pls show a starting frame and well as pd.show_versions()

[691175002]

Versions:

INSTALLED VERSIONS
------------------
commit: None
python: 3.5.2.final.0
python-bits: 64
OS: Windows
OS-release: 7
machine: AMD64
processor: Intel64 Family 6 Model 60 Stepping 3, GenuineIntel
byteorder: little
LC_ALL: None
LANG: en
LOCALE: None.None

pandas: 0.19.0
nose: None
pip: 8.1.2
setuptools: 27.2.0
Cython: None
numpy: 1.11.2
scipy: 0.18.1
statsmodels: 0.8.0rc1
xarray: None
IPython: 5.1.0
sphinx: 1.4.8
patsy: 0.4.1
dateutil: 2.5.3
pytz: 2016.7
blosc: None
bottleneck: None
tables: None
numexpr: None
matplotlib: 1.5.3
openpyxl: 2.4.0
xlrd: 1.0.0
xlwt: None
xlsxwriter: 0.9.3
lxml: None
bs4: 4.5.1
html5lib: None
httplib2: None
apiclient: None
sqlalchemy: None
pymysql: None
psycopg2: None
jinja2: 2.8
boto: None
pandas_datareader: None

Easy reproduction:

arrays = [np.array(['bar', 'bar', 'baz', 'baz', 'foo', 'foo', 'qux', 'qux']),
          np.array(['one', 'two', 'one', 'two', 'one', 'two', 'one', 'two'])]
df = pd.DataFrame(np.random.randn(8, 2), index=arrays)

df.groupby(level=0).rolling(2).sum()
Out[17]: 
                    0         1
bar bar one       NaN       NaN
        two -0.779420 -1.860285
baz baz one       NaN       NaN
        two -0.680321 -1.221606
foo foo one       NaN       NaN
        two -0.992899 -0.612821
qux qux one       NaN       NaN
        two -1.694832 -1.078676

df.groupby(level=0).apply(lambda x: x.rolling(2).sum())
Out[18]: 
                0         1
bar one       NaN       NaN
    two -0.779420 -1.860285
baz one       NaN       NaN
    two -0.680321 -1.221606
foo one       NaN       NaN
    two -0.992899 -0.612821
qux one       NaN       NaN
    two -1.694832 -1.078676

[jreback]

duplicate of #14013

pull-requests are welcome to fix!