Calendar

You can use Calendar object in MCQuantLib to reform your pricing. It is based on QuantLib. To use it, just import QuantLib and import Calendar from MCQuantLib:

import QuantLib as ql
from MCQuantLib import Calendar

Calendar accepts a QuantLib calendar instance as its input. You can use any calendar object that QuantLib provides, such as:

jp = Calendar(ql.Japan())
uk = Calendar(ql.UnitedKingdom())

Add Holiday

Use addHolidays to add some holiday so that it can suit your condition. This function accepts a collection of pandas.Timestamp:

jp.addHolidays([pd.Timestamp('20250104'), pd.Timestamp('20250204')])

Trading Days

Most useful function of Calendar instance is isTrading, it returns True or False, after given a pandas.Timestamp:

jp.isTrading(pd.Timestamp('20250104'))

To generate a trading days list between given start day and end day, use:

jp.tradingDaysBetween(start=pd.Timestamp('20250104'), end=pd.Timestamp('20250204'))

To get the next trading day, or last trading day, or trading day in one month later, or any period later or before, you can use offset, like:

jp.offset(date=pd.Timestamp('20250104'), n=1, period=ql.Days)  # get the next trading day
jp.offset(date=pd.Timestamp('20250104'), n=3, period=ql.Days)  # get 3 trading days later
jp.offset(date=pd.Timestamp('20250104'), n=1, period=ql.Months)  # get the trading day after 1 month
jp.offset(date=pd.Timestamp('20250104'), n=1, period=ql.Years)  # get the trading day after 1 year
jp.offset(date=pd.Timestamp('20250104'), n=-16, period=ql.Weeks)  # get the trading day before 16 weeks

To calculate how many trading days exist between start day and end day, use:

jp.numTradingDaysBetween(start=pd.Timestamp('20250104'), end=pd.Timestamp('20250204'))

To calculate how many trading days exist between start day and a list of target day, use:

jp.numTradingDaysBetweenGrid(start=pd.Timestamp('20250104'), dateList=[pd.Timestamp('20250204'), pd.Timestamp('20250304')])

Schedule

To generate a trading day list between start and end day, with a given step, use:

jp.makeSchedule(start=pd.Timestamp('20250104'), end=pd.Timestamp('20250704'), period='3w')

You can use a string to specify period, such as: 1d, 18d, 3w, 2m, 2y.

This function also has parameter named forwardAdjust and forwardScale. If forwardAdjust is True, then the day will be adjusted to next trading day if it is not a trading day. If False, it will be adjusted to last trading day.

If forwardScale is True, then the last period of generated list could be shorter than given. If False, then the first period of generated list could be shorter than given

To generate a trading day list given period and number of period, use:

jp.makeScheduleByPeriod(start=pd.Timestamp('20250104'), period='3w', count=5)

It also accepts a bool as parameter named as forwardAdjust, if True, a day will be adjusted into next trading day if it is not a trading day, else it will be adjusted to last trading day.

Self-Defined Calendar

If you want to define your own calendar, you just need to initialize a Calendar object with ql.NullCalendar() and add your own holidays:

import QuantLib as ql
myCalendar = Calendar(ql.NullCalendar())
myCalendar.addHolidays([pd.Timestamp('20250104'), pd.Timestamp('20250105')])

It is so crazy that you only rest rwo days in a year, but anyway, you have your own calendar now.