Find start date of previous 12 month from current date in python-CodePudding

Code

today = pd.to_datetime('today').strftime("%d/%m/%Y")
last = (pd.to_datetime(today) - pd.DateOffset(years=1)).strftime("%d/%m/%Y")

listofmonths = pd.date_range(start=last, end=today, freq='MS') 
listofmonths

Output:

DatetimeIndex(['2021-06-01', '2021-07-01', '2021-08-01', '2021-09-01',
           '2021-10-01', '2021-11-01', '2021-12-01', '2022-01-01'],
          dtype='datetime64[ns]', freq='MS')

Note: Refer the output in which i am getting only 8 months and i want past 12 months first date from the current date

Expected Output:

 **DatetimeIndex(['2021-06-01', '2021-07-01', '2021-08-01', '2021-09-01',
           '2021-10-01', '2021-11-01', '2021-12-01','2021-01-01', '2022-02-01',
           '2022-03-01','2022-04-01', '2022-05-01', '2022-06-01',],
          dtype='datetime64[ns]', freq='MS')**

CodePudding user response：

Try to update strftime from "%d/%m/%Y" to "%m/%d/%Y":

today = pd.to_datetime('today').strftime("%m/%d/%Y")
last = (pd.to_datetime(today) - pd.DateOffset(years=1)).strftime("%m/%d/%Y")

listofmonths = pd.date_range(start=last, end=today, freq='MS')

Output:

DatetimeIndex(['2021-06-01', '2021-07-01', '2021-08-01', '2021-09-01',
               '2021-10-01', '2021-11-01', '2021-12-01', '2022-01-01',
               '2022-02-01', '2022-03-01', '2022-04-01', '2022-05-01',
               '2022-06-01'],
              dtype='datetime64[ns]', freq='MS')

P.s. the issue was, that 'today' was treated as January, 6:

>>> last
'06/01/2021'
>>> today
'01/06/2022'

CodePudding user response：

Problem is that %d/%m/%Y is dayfirst however by default pandas use month first

today = pd.to_datetime('today')
last = (pd.to_datetime(today) - pd.DateOffset(years=1))

listofmonths = pd.date_range(start=last, end=today, freq='MS').strftime("%d/%m/%Y")

Index(['01/06/2021', '01/07/2021', '01/08/2021', '01/09/2021', '01/10/2021',
       '01/11/2021', '01/12/2021', '01/01/2022', '01/02/2022', '01/03/2022',
       '01/04/2022', '01/05/2022', '01/06/2022'],
      dtype='object')

CodePudding user response：

It'd be more advisable to use datetime/timestamp object for the pd.date_range instead of string.

today_ts = pd.to_datetime('today').floor(freq='D')
last_ts = (pd.to_datetime(today_ts) - pd.DateOffset(years=1))

listofmonths = pd.date_range(start=last_ts, end=today_ts, freq='MS') 
listofmonths
DatetimeIndex(['2021-06-01', '2021-07-01', '2021-08-01', '2021-09-01',
               '2021-10-01', '2021-11-01', '2021-12-01', '2022-01-01',
               '2022-02-01', '2022-03-01', '2022-04-01', '2022-05-01',
               '2022-06-01'],
              dtype='datetime64[ns]', freq='MS')

And if you need to use them as string, you may convert it later

CodePudding user response：

Try using standard datetime python library for anything datetime related.
You error might have been a type error.
This code works fine for me.

from datetime import timedelta, datetime

today = datetime.now().strftime("%Y-%m-%d")
past = (datetime.now() - timedelta(days=365)).strftime("%Y-%m-%d")

pd.date_range(past, today, freq="MS")

Output:

DatetimeIndex(['2021-06-01', '2021-07-01', '2021-08-01', '2021-09-01',
               '2021-10-01', '2021-11-01', '2021-12-01', '2022-01-01',
               '2022-02-01', '2022-03-01', '2022-04-01', '2022-05-01',
               '2022-06-01'],
              dtype='datetime64[ns]', freq='MS')