BUG: QuarterBegin Does not work with Period

[cancan101]

Pandas version checks

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of pandas.

• I have confirmed this bug exists on the main branch of pandas.

Reproducible Example

import pandas as pd
pd.Period(year=2014, quarter=3, freq=pd.offsets.QuarterBegin(startingMonth=12))

Issue Description

Unable to to create a Quarterly period as defined by the start of the quarter vs the end.

Might be related to this very old issue: #5091

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
File period.pyx:2839, in pandas._libs.tslibs.period.freq_to_dtype_code()

AttributeError: 'pandas._libs.tslibs.offsets.QuarterBegin' object has no attribute '_period_dtype_code'

The above exception was the direct cause of the following exception:

ValueError                                Traceback (most recent call last)
File period.pyx:2841, in pandas._libs.tslibs.period.freq_to_dtype_code()

ValueError: Invalid frequency: <QuarterBegin: startingMonth=12>
Exception ignored in: 'pandas._libs.tslibs.period._ordinal_from_fields'
Traceback (most recent call last):
  File "period.pyx", line 2841, in pandas._libs.tslibs.period.freq_to_dtype_code
ValueError: Invalid frequency: <QuarterBegin: startingMonth=12>
---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
Cell In[14], line 1
----> 1 pd.Period(year=2014, quarter=3, freq=pd.offsets.QuarterBegin(startingMonth=12))

File period.pyx:2830, in pandas._libs.tslibs.period.Period.__new__()

File period.pyx:1779, in pandas._libs.tslibs.period._Period._from_ordinal()

File period.pyx:1742, in pandas._libs.tslibs.period._Period.__cinit__()

AttributeError: 'pandas._libs.tslibs.offsets.QuarterBegin' object has no attribute '_period_dtype_code'

Expected Behavior

Period('2014Q3', 'QS-DEC')

Installed Versions

INSTALLED VERSIONS

commit : a60ad39
python : 3.12.0.final.0
python-bits : 64
OS : Darwin
OS-release : 22.1.0
Version : Darwin Kernel Version 22.1.0: Sun Oct 9 20:19:12 PDT 2022; root:xnu-8792.41.9~3/RELEASE_ARM64_T6020
machine : arm64
processor : arm
byteorder : little
LC_ALL : None
LANG : en_US.UTF-8
LOCALE : en_US.UTF-8

pandas : 2.1.2
numpy : 1.26.1
pytz : 2023.3.post1
dateutil : 2.8.2
setuptools : 68.2.2
pip : 23.3.1
Cython : None
pytest : None
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : None
lxml.etree : None
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : 3.1.2
IPython : 8.17.2
pandas_datareader : None
bs4 : 4.12.2
bottleneck : None
dataframe-api-compat: None
fastparquet : None
fsspec : None
gcsfs : None
matplotlib : 3.8.0
numba : None
numexpr : None
odfpy : None
openpyxl : None
pandas_gbq : None
pyarrow : None
pyreadstat : None
pyxlsb : None
s3fs : None
scipy : 1.11.3
sqlalchemy : None
tables : None
tabulate : None
xarray : None
xlrd : None
zstandard : None
tzdata : 2023.3
qtpy : None
pyqt5 : None

[hvsesha]

@cancan101
Try this for quarter This is working

import pandas as pd
period=pd.Period(year=2014, month=3, freq="Q")

print (period)

[cancan101]

That is the same as using QuarterEnd. These are the same:

pd.Period(year=2014, month=3, freq="Q")

and

pd.Period(year=2014, quarter=3, freq=pd.offsets.QuarterEnd(startingMonth=12))

However, I want to use QuarterBegin.

[Dhayanidhi-M]

Hi ,

the above error says , "_period_dtype_code" attribute not found . Which means the class "QuarterBegin" does not have the attribute "_period_dtype_code" .

I checked the code , in file "offset.pyx" , we have the class "QuarterBegin" and it does not have the attribute "_period_dtype_code" . Hence we are seeing the above error.

I think , this needs to be fixed in code . Is someone working on this ?.

[cancan101]

Yea it seems like that code should be fixed with either support or a better error message. and / or the docs might need to be made more clear as the way I read them, QuarterBegin should work:

where the S is an optional modification to the alias string.

[jbrockmendel]

cc @natmokval maybe we should avoid having _period_dtype_code bubble up in the exception message? we need to be more clear that some offsets are not supported as period freqs and that this is intentional

[natmokval]

Thank you @cancan101 for the report.

We are working on renaming some offset aliases, e.g. 'M' to 'ME' (#52064), 'Q' to 'QE' ( #55553), 'Y' to 'YE' (#55792). The goal is to separate them from Period aliases: 'M', 'Q', 'Y'. These changes are on main so far and they will be in the next release.

Expected Behavior is Period('2014Q3', 'Q-DEC') and it doesn't suppose to be Period('2014Q3', 'QS-DEC') or Period('2014Q3', 'QE-DEC').

a valid frequency for Period can be either string 'Q' or offsets class pd.offsets.QuarterEnd

[natmokval]

That is the same as using QuarterEnd. These are the same:

pd.Period(year=2014, month=3, freq="Q")

and

pd.Period(year=2014, quarter=3, freq=pd.offsets.QuarterEnd(startingMonth=12))

However, I want to use QuarterBegin.

some offsets are not supported as period frequencies, e.g. offset class QuarterBegin and corresponding string 'QS' are not valid for Period.

[natmokval]

Yea it seems like that code should be fixed with either support or a better error message. and / or the docs might need to be made more clear as the way I read them, QuarterBegin should work: where the S is an optional modification to the alias string.

Thank you, @cancan101. I agree, we need a better error message and documentation might be more clear in this part. I think we should show an error message that some offsets are not supported as period frequencies and I will make PR for this.

Documentation for Time series is updated (link) . In the new version we point out that for Anchored offsets only frequencies like "(B)Q(E)(S)\-DEC" can be used:

"(B)Q(E)(S)\-DEC", "quarterly frequency, year ends in December. Same as 'QE'" 

[natmokval]

cc @natmokval maybe we should avoid having _period_dtype_code bubble up in the exception message? we need to be more clear that some offsets are not supported as period freqs and that this is intentional

thank you, I agree, it would be better to raise a ValueError, with a message that some offsets are not supported as period frequencies. I will make PR.
cc @MarcoGorelli