Holidays

In [1]:

import pandas as pd
df = pd.read_csv("apple_stock_price_nodate.csv")
df.head()

Out[1]:

	Open	High	Low	Close	Adj Close	Volume
0	190.679993	191.960007	189.559998	191.610001	188.737030	15989400
1	192.449997	193.660004	192.050003	193.000000	190.106216	18697900
2	193.059998	194.850006	192.429993	194.820007	191.898926	16709900
3	194.610001	195.960007	193.610001	194.210007	191.298080	19076000
4	194.990005	195.190002	190.100006	190.979996	188.116501	24024000

Using 'B' frequency is not going to help because 4th July was holiday and 'B' is not taking that into account. It only accounts for weekends

Generate US holidays calendar frequency

In [2]:

from pandas.tseries.holiday import USFederalHolidayCalendar
from pandas.tseries.offsets import CustomBusinessDay

us_cal = CustomBusinessDay(calendar=USFederalHolidayCalendar())

rng = pd.date_range(start="2018-07-23",periods=df.shape[0],freq=us_cal)
rng

Out[2]:

DatetimeIndex(['2018-07-23', '2018-07-24', '2018-07-25', '2018-07-26',
               '2018-07-27', '2018-07-30', '2018-07-31', '2018-08-01',
               '2018-08-02', '2018-08-03',
               ...
               '2019-07-10', '2019-07-11', '2019-07-12', '2019-07-15',
               '2019-07-16', '2019-07-17', '2019-07-18', '2019-07-19',
               '2019-07-22', '2019-07-23'],
              dtype='datetime64[ns]', length=252, freq='C')

In [5]:

df.set_index(rng,inplace=True)
df.head()

Out[5]:

	Open	High	Low	Close	Adj Close	Volume
2018-07-23	190.679993	191.960007	189.559998	191.610001	188.737030	15989400
2018-07-24	192.449997	193.660004	192.050003	193.000000	190.106216	18697900
2018-07-25	193.059998	194.850006	192.429993	194.820007	191.898926	16709900
2018-07-26	194.610001	195.960007	193.610001	194.210007	191.298080	19076000
2018-07-27	194.990005	195.190002	190.100006	190.979996	188.116501	24024000

You can define your own calendar using AbstractHolidayCalendar as shown below. USFederalHolidayCalendar is the only calendar available in pandas library and it serves as an example for those who want to write their own custom calendars. Here is the link for USFederalHolidayCalendar implementation https://github.com/pandas-dev/pandas/blob/master/pandas/tseries/holiday.py

In [12]:

from pandas.tseries.holiday import AbstractHolidayCalendar, nearest_workday, Holiday
class myCalendar(AbstractHolidayCalendar):
    rules = [
        Holiday('My Birth Day', month=5, day=13),#, observance=nearest_workday),
    ]
    
my_bday = CustomBusinessDay(calendar=myCalendar())

pd.date_range('5/1/2019','6/1/2019',freq=my_bday)

Out[12]:

DatetimeIndex(['2019-05-01', '2019-05-02', '2019-05-03', '2019-05-06',
               '2019-05-07', '2019-05-08', '2019-05-09', '2019-05-10',
               '2019-05-14', '2019-05-15', '2019-05-16', '2019-05-17',
               '2019-05-20', '2019-05-21', '2019-05-22', '2019-05-23',
               '2019-05-24', '2019-05-27', '2019-05-28', '2019-05-29',
               '2019-05-30', '2019-05-31'],
              dtype='datetime64[ns]', freq='C')

Weekend in egypt is Friday and Saturday. Sunday is just a normal weekday and you can handle this custom week schedule using CystomBysinessDay with weekmask as shown below

In [15]:

egypt_weekdays = "Sun Mon Tue Wed Thu"

b = CustomBusinessDay(weekmask=egypt_weekdays)

pd.date_range(start="7/1/2018",periods=20,freq=b)

Out[15]:

DatetimeIndex(['2018-07-01', '2018-07-02', '2018-07-03', '2018-07-04',
               '2018-07-05', '2018-07-08', '2018-07-09', '2018-07-10',
               '2018-07-11', '2018-07-12', '2018-07-15', '2018-07-16',
               '2018-07-17', '2018-07-18', '2018-07-19', '2018-07-22',
               '2018-07-23', '2018-07-24', '2018-07-25', '2018-07-26'],
              dtype='datetime64[ns]', freq='C')

You can also add holidays to this custom business day frequency¶

In [17]:

b = CustomBusinessDay(holidays=['2018-08-04', '2018-08-05'], weekmask=egypt_weekdays)

pd.date_range(start="8/1/2018",periods=20,freq=b)

Out[17]:

DatetimeIndex(['2018-08-01', '2018-08-02', '2018-08-06', '2018-08-07',
               '2018-08-08', '2018-08-09', '2018-08-12', '2018-08-13',
               '2018-08-14', '2018-08-15', '2018-08-16', '2018-08-19',
               '2018-08-20', '2018-08-21', '2018-08-22', '2018-08-23',
               '2018-08-26', '2018-08-27', '2018-08-28', '2018-08-29'],
              dtype='datetime64[ns]', freq='C')

Mathematical operations on date object using custom business day¶

In [18]:

from datetime import datetime
dt = datetime(2017,7,9)
dt

Out[18]:

datetime.datetime(2017, 7, 9, 0, 0)

Out[19]:

Timestamp('2017-07-10 00:00:00')