Skip to main content

Functions to handle business days calculations

Project description

bizdays computes business days between two dates based on the definition of nonworking days (usually holidays and weekends—nonworking weekdays). It also computes other collateral effects like adjust dates for the next or previous business day, check whether a date is a business day, create generators of business days sequences, and so forth.


bizdays is avalilable at PyPI, so it is pip instalable.

pip install bizdays


Business days calculations are done defining a Calendar object.

from bizdays import Calendar
cal = Calendar(holidays, ['Sunday', 'Saturday'])

where holidays is a sequence of dates which represents nonworking dates and the second argument, weekdays, is a sequence with nonworking weekdays. holidays must be a sequence of strings with ISO formatted dates or objects and weekdays a sequence of weekdays in words.

Once you have a Calendar you can

>>> cal.isbizday('2014-01-12')
>>> cal.isbizday('2014-01-13')
>>> cal.bizdays('2014-01-13', '2015-01-13')
>>> cal.adjust_next('2015-12-25'), 12, 28)
>>> cal.adjust_next('2015-12-28'), 12, 28)
>>> cal.adjust_previous('2014-01-01'), 12, 31)
>>> cal.adjust_previous('2014-01-02'), 1, 2)
>>> cal.seq('2014-01-02', '2014-01-07')
<generator object seq at 0x1092b02d0>
>>> list(cal.seq('2014-01-02', '2014-01-07'))
[, 1, 2),, 1, 3),, 1, 6),, 1, 7)]
>>> cal.offset('2014-01-02', 5), 1, 9)
>>> cal.getdate('15th day', 2002, 5), 5, 15)
>>> cal.getdate('15th bizday', 2002, 5), 5, 22)
>>> cal.getdate('last wed', 2002, 5), 5, 29)
>>> cal.getdate('first fri before last day ', 2002, 5), 5, 24)

In this example I used the list of holidays released by ANBIMA.

Important note on date arguments and returning dates

As you can see in the examples all date arguments are strings ISO formatted (YYYY-mm-dd or %Y-%m-%d), but they can also be passed as objects. All returning dates are objects (or a sequence of it), unless you set iso=True, that will return an ISO formatted string.

The startdate and enddate of a Calendar are defined accordingly the first and last given holidays.


To compute the business days between two dates you call bizdays passing from and to dates as arguments.

>>> cal.bizdays('2012-12-31', '2013-01-03')


You specify dates by its position or related to other dates, for example:

>>> cal.getdate('15th day', 2002, 5), 5, 15)

it returns the 15th day of 2002 may. You can also reffer to the whole year.

>>> cal.getdate('150th day', 2002), 5, 30)

It accepts day, bizday and weekdays by: sun, mon, tue, wed, thu, fri, and sat.

>>> cal.getdate('last day', 2006), 12, 31)
>>> cal.getdate('last bizday', 2006), 12, 29)
>>> cal.getdate('last mon', 2006), 12, 25)

For postion use: first, second, third, 1st, 2nd, 3rd, [n]th, and last.

Using date postions as a reference

You can find before and after other date positions (using date positions as a reference).

>>> cal.getdate('last mon before 30th day', 2006, 7), 7, 24)
>>> cal.getdate('second bizday after 15th day', 2006), 1, 18)

following and preceding

Several contracts, by default, always expiry in the same day, for example, 1st Januray, which isn't a business day, so instead of carrying your code with awful checks you could call following which returns the given date whether it is a business day or the next business day.

>>> cal.following('2013-01-01'), 1, 2)
>>> cal.following('2013-01-02'), 1, 2)

We also have preceding, although I suppose it is unusual, too.

>>> cal.preceding('2013-01-01'), 12, 31)

modified_following and modified_preceding

modified_following and modified_preceding are common functions used to specify maturity of contracts. They work the same way following and preceding but once the returning date is a different month it is adjusted to the following or preceding business day in the same month.

>>> dt = cal.getdate('last day', 2002, 3)
>>> dt, 3, 31)
>>> cal.modified_following(dt, iso=True)
>>> cal.isbizday('2002-03-29')
>>> dt = cal.getdate('first day', 2002, 6)
>>> dt, 6, 1)
>>> cal.modified_preceding(dt, iso=True)


To execute calculations through sequential dates, sometimes you must consider only business days. For example, you want to compute the price of a bond from its issue date up to its maturity. You have to walk over business days in order to carry the contract up to maturity. To accomplish that you use the seq method (stolen from R) which returns a sequence generator of business days.

>>> for dt in cal.seq('2012-12-31', '2013-01-03'):
...     print dt


This method offsets the given date by n days respecting the calendar, so it obligatorily returns a business day.

>>> cal.offset('2013-01-02', 1), 1, 3)
>>> cal.offset('2013-01-02', 3), 1, 7)
>>> cal.offset('2013-01-02', 0), 1, 2)

Obviously, if you want to offset backwards you can use -n.

>>> print cal.offset('2013-01-02', -1)
>>> print cal.offset('2013-01-02', -3)

Once the given date is a business day there is no problems, but if instead it isn't a working day the offset can lead to unexpected results. For example:

>>> cal.offset('2013-01-01', 1), 1, 2)
>>> cal.offset('2013-01-01', 0), 1, 1)
>>> cal.offset('2013-01-01', -1), 12, 31)

Actual Calendar

The Actual Calendar can be defined as

>>> cal = Calendar(name='actual')
>>> cal
Calendar: actual
Start: 1970-01-01
End: 2071-01-01
Holidays: 0
Financial: True

The Actual Calendar doesn't consider holidays, nor nonworking weekdays for counting business days between 2 dates. This is the same of subtracting 2 dates, and adjust methods will return the given argument. But the idea of using the Actual Calendar is working with the same interface for any calendar you work with. When you price financial instruments you don't have to check if it uses business days or not.

startdate and enddate defaults to 1970-01-01 and 2071-01-01, but they can be set during Calendar's instanciation.

Vectorized operations

The Calendar's methods: isbizday, bizdays, adjust_previous, adjust_next, and offset, have a vectorized counterparty, inside Calendar.vec attribute.

>>> cal = Calendar.load('')
>>> dates = ('2002-01-01', '2002-01-02', '2002-01-03')
>>> tuple(cal.vec.adjust_next(dates))
(, 1, 2),, 1, 2),, 1, 3))
>>> list(cal.vec.bizdays('2001-12-31', dates))
[0, 1, 2]

These functions accept sequences and single values, but always return generators. In bizdays call a date and a sequence have been passed, computing business days between that date and all the others.

Recycle rule

Once you pass 2 sequences for bizdays and offset and those sequences doesn't have the same length, no problem. The shorter collection is cycled to fit the longer's length.

>>> dates = ('2002-01-01', '2002-01-02', '2002-01-03', '2002-01-04', '2002-01-05')
>>> tuple(cal.vec.offset(dates, (1, 2, 3)))
(, 1, 3),, 1, 4),, 1, 8),, 1, 7),, 1, 9))

These methods work well with sequences but not with generators, since I haven't found an easy way to find out which generator is the shorter.

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for bizdays, version 0.3.0
Filename, size File type Python version Upload date Hashes
Filename, size bizdays-0.3.0-py3-none-any.whl (8.6 kB) File type Wheel Python version py3 Upload date Hashes View
Filename, size bizdays-0.3.0.tar.gz (8.5 kB) File type Source Python version None Upload date Hashes View

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring DigiCert DigiCert EV certificate Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page