pandas Integration

bizdays has a very handy integration with pandas DataFrame.

[1]:

from bizdays import set_option, Calendar

set_option('mode', 'pandas')

When the option mode is set to pandas the calendar methods will return

  • Timestamp instead of datetime.date

  • DatetimeIndex instead of lists of dates

  • numpy.ndarray instead of lists of integers (bizdays) and bools (isbizday)

[2]:

cal = Calendar.load(filename='ANBIMA.cal')
cal

[2]:

Calendar: ANBIMA
Start: 2000-01-01
End: 2078-12-25
Weekdays: Saturday, Sunday
Holidays: 948
Financial: True

[3]:

cal.isbizday(['2014-01-12', '2014-01-13'])

[3]:

array([False,  True])

[4]:

cal.bizdays('2014-01-13', ['2015-01-13', '2016-01-13', None])

[4]:

array([253, 503, <NA>], dtype=object)

[5]:

cal.following('2015-12-25')

[5]:

Timestamp('2015-12-28 00:00:00')

[6]:

cal.following(['2015-12-28', '2015-12-25'])

[6]:

DatetimeIndex(['2015-12-28', '2015-12-28'], dtype='datetime64[ns]', freq=None)

[7]:

cal.seq('2014-01-02', '2014-01-07')

[7]:

DatetimeIndex(['2014-01-02', '2014-01-03', '2014-01-06', '2014-01-07'], dtype='datetime64[ns]', freq=None)

[8]:

cal.offset('2014-01-02', range(10))

[8]:

DatetimeIndex(['2014-01-02', '2014-01-03', '2014-01-06', '2014-01-07',
               '2014-01-08', '2014-01-09', '2014-01-10', '2014-01-13',
               '2014-01-14', '2014-01-15'],
              dtype='datetime64[ns]', freq=None)

[9]:

cal.getdate('15th bizday', 2002, 5)

[9]:

Timestamp('2002-05-22 00:00:00')

[10]:

cal.getdate('15th day', 2002, range(1, 13))

[10]:

DatetimeIndex(['2002-01-15', '2002-02-15', '2002-03-15', '2002-04-15',
               '2002-05-15', '2002-06-15', '2002-07-15', '2002-08-15',
               '2002-09-15', '2002-10-15', '2002-11-15', '2002-12-15'],
              dtype='datetime64[ns]', freq=None)

[11]:

cal.getbizdays(2021, range(1, 13))

[11]:

array([20, 18, 23, 20, 21, 21, 22, 22, 21, 20, 20, 23])

Example: series of payments

Let’s create a series of payment dates and compute the business days between each payment date.

The payments are scheduled to occur in the 15th days of each month of 2021.

Start creating a DataFrame with year and month columns.

[12]:

import pandas as pd

df = pd.DataFrame({
    'year': 2021,
    'month': range(1,13)
})
df

[12]:
year month
0 2021 1
1 2021 2
2 2021 3
3 2021 4
4 2021 5
5 2021 6
6 2021 7
7 2021 8
8 2021 9
9 2021 10
10 2021 11
11 2021 12

Use getdate to get the payment dates.

[13]:

df['payment_dates'] = cal.getdate('15th day', df['year'], df['month'])
df

[13]:
year month payment_dates
0 2021 1 2021-01-15
1 2021 2 2021-02-15
2 2021 3 2021-03-15
3 2021 4 2021-04-15
4 2021 5 2021-05-15
5 2021 6 2021-06-15
6 2021 7 2021-07-15
7 2021 8 2021-08-15
8 2021 9 2021-09-15
9 2021 10 2021-10-15
10 2021 11 2021-11-15
11 2021 12 2021-12-15

The payments happen in business days so the dates must be shifted to the next working date.

[14]:

df['fixing'] = cal.following(df['payment_dates'])
df

[14]:
year month payment_dates fixing
0 2021 1 2021-01-15 2021-01-15
1 2021 2 2021-02-15 2021-02-17
2 2021 3 2021-03-15 2021-03-15
3 2021 4 2021-04-15 2021-04-15
4 2021 5 2021-05-15 2021-05-17
5 2021 6 2021-06-15 2021-06-15
6 2021 7 2021-07-15 2021-07-15
7 2021 8 2021-08-15 2021-08-16
8 2021 9 2021-09-15 2021-09-15
9 2021 10 2021-10-15 2021-10-15
10 2021 11 2021-11-15 2021-11-16
11 2021 12 2021-12-15 2021-12-15

bizdays compute the amount of business days between each payment.

[15]:

df['business_days'] = cal.bizdays(df['fixing'].shift(), df['fixing'])
df

[15]:
year month payment_dates fixing business_days
0 2021 1 2021-01-15 2021-01-15 <NA>
1 2021 2 2021-02-15 2021-02-17 21
2 2021 3 2021-03-15 2021-03-15 18
3 2021 4 2021-04-15 2021-04-15 22
4 2021 5 2021-05-15 2021-05-17 21
5 2021 6 2021-06-15 2021-06-15 20
6 2021 7 2021-07-15 2021-07-15 22
7 2021 8 2021-08-15 2021-08-16 22
8 2021 9 2021-09-15 2021-09-15 21
9 2021 10 2021-10-15 2021-10-15 21
10 2021 11 2021-11-15 2021-11-16 20
11 2021 12 2021-12-15 2021-12-15 21