DOC: Anchored Offset Starting Quarter and Year defaults to JAN not DEC

[tdpetrou]

Code Sample, a copy-pastable example if possible

>>> index = pd.date_range('2012-1-1', '2013-12-31')
>>> np.random.seed(1)
>>> data = np.random.randint(0, 10, len(index))
>>> s = pd.Series(index=index, data=data)

>>> s.resample('Q').sum()
2012-03-31    457
2012-06-30    404
2012-09-30    401
2012-12-31    415
2013-03-31    383
2013-06-30    382
2013-09-30    398
2013-12-31    387
Freq: Q-DEC, dtype: int64

>>> s.resample('Q-DEC').sum() # same as above

>>> s.resample('QS').sum() # defaults to QS-JAN
2012-01-01    457
2012-04-01    404
2012-07-01    401
2012-10-01    415
2013-01-01    383
2013-04-01    382
2013-07-01    398
2013-10-01    387
Freq: QS-JAN, dtype: int64

>>> s.resample('QS-DEC').sum()
2011-12-01    299
2012-03-01    427
2012-06-01    381
2012-09-01    438
2012-12-01    386
2013-03-01    396
2013-06-01    375
2013-09-01    414
2013-12-01    111
Freq: QS-DEC, dtype: int64

Problem description

The docs have

(B)Q(S)-DEC	quarterly frequency, year ends in December. Same as ‘Q’

But QS is the same as QS-JAN not December. The same thing for AS.

[jreback]

its not if you see this issue: #5307

[tdpetrou]

I'm confused. That issue doesn't appear to have anything to do with this. The docs seem clearly wrong. The docs inform the user that QS defaults to QS-DEC. It does not. It defaults to QS-JAN.