Derivatives, fixed income & risk
Calendars for derivatives pricing, valuation and risk
Swaps, swaptions, bonds, structured notes and loans carry cash-flow schedules that run for decades. Every accrual, fixing and payment date is adjusted on a joint calendar with a business-day convention. A wrong holiday moves a payment date, changes the accrual and the discount factor, and breaks the match with your counterparty or the clearing house. TradingCalendars provides the calendars, the conventions and the schedule itself.
Who uses it
- Quant and pricing teams at banks and funds
- Valuation and risk teams
- Fintech and analytics vendors building pricing libraries
- Loan, bond and structured-product platforms
Where schedules go wrong
Hard-coded calendars that age
Holidays change: Juneteenth became a US federal holiday in 2021, and the UK added one-off bank holidays for the State Funeral (19 September 2022) and the Coronation (8 May 2023). A calendar shipped in a library release stays wrong until the next upgrade.
Calendars that run out
A 30-year swap traded today needs holidays into the 2050s. Our calendars are generated from rules through 2077, and every answer says how far the data goes and flags estimated dates.
The wrong calendar for the leg
Payment dates usually follow the business centres in the confirmation (New York, London, TARGET), while the rate fixing follows the index's own calendar. The schedule endpoint takes separate payment and fixing calendars.
Convention edge cases
Modified following at month end, the end-of-month rule, short and long stubs, payment lags: small differences that change accrual days and cash flows.
Live example
The next non-business days under each convention, New York + London
| Date | Closed for | Following | Modified following | Preceding | Modified preceding |
|---|---|---|---|---|---|
| Thu, 26 Nov 2026 | Thanksgiving Day (NY) | Fri 27 Nov | Fri 27 Nov | Wed 25 Nov | Wed 25 Nov |
| Fri, 25 Dec 2026 | Christmas Day (NY), Christmas Day (London) | Tue 29 Dec | Tue 29 Dec | Thu 24 Dec | Thu 24 Dec |
| Mon, 28 Dec 2026 | Boxing Day (London) | Tue 29 Dec | Tue 29 Dec | Thu 24 Dec | Thu 24 Dec |
| Fri, 1 Jan 2027 | New Year's Day (NY), New Year's Day (London) | Mon 4 Jan | Mon 4 Jan | Thu 31 Dec (previous month) | Mon 4 Jan |
| Mon, 18 Jan 2027 | Birthday of Martin Luther King, Jr. (NY) | Tue 19 Jan | Tue 19 Jan | Fri 15 Jan | Fri 15 Jan |
| Sun, 31 Jan 2027 | weekend | Mon 1 Feb (next month) | Fri 29 Jan | Fri 29 Jan | Fri 29 Jan |
Upcoming holidays and month-end weekends on the joint calendar of USNYC.BANK and GBLON.BANK, and where each ISDA business day convention moves a payment date that falls on them. Modified conventions never leave the month.
Live example
A 1-year quarterly schedule starting at spot today, New York + London
| Period | Unadjusted end | Adjusted end | Accrual days | Why it moved |
|---|---|---|---|---|
| 1 | Fri, 19 Feb 2027 | Fri, 19 Feb 2027 | 92 | - |
| 2 | Wed, 19 May 2027 | Wed, 19 May 2027 | 89 | - |
| 3 | Thu, 19 Aug 2027 | Thu, 19 Aug 2027 | 92 | - |
| 4 | Fri, 19 Nov 2027 | Fri, 19 Nov 2027 | 92 | - |
Start = T+2 on the joint calendar; quarterly periods from the start date; every date adjusted with modified following on USNYC.BANK and GBLON.BANK. The /v1/schedule endpoint does this for up to 60 years, with stubs, end-of-month, payment and fixing lags.
What you get
Business-day conventions
Following, modified following, preceding and modified preceding (ISDA), on one calendar or a joint calendar.
Schedules up to 60 years
Accrual periods with stubs and the end-of-month rule, payment dates with a lag, and fixing dates on their own calendar.
Into your library
Load closure dates into QuantLib, pandas or an in-house library, or call the schedule endpoint directly.
Calendars to start with
Use it your way
The REST API, the open-source Python client, or your AI assistant through the MCP server. Full reference in the documentation.
curl "https://api.tradingcalendars.com/v1/schedule?calendars=USNYC.BANK,GBLON.BANK&start=2026-11-30&end=2036-11-30&frequency=6M&convention=modified_following&end_of_month=true&payment_lag=2" \
-H "Authorization: Bearer YOUR_API_KEY"
curl "https://api.tradingcalendars.com/v1/adjust-date?calendars=USNYC.BANK,GBLON.BANK&date=2027-05-29&convention=modified_following" \
-H "Authorization: Bearer YOUR_API_KEY"import QuantLib as ql
from tradingcalendars import TradingCalendars
tc = TradingCalendars()
# A joint New York + London calendar for QuantLib, through 2077
cal = ql.BespokeCalendar("NY+LDN")
cal.addWeekend(ql.Saturday)
cal.addWeekend(ql.Sunday)
for code in ("USNYC.BANK", "GBLON.BANK"):
for d in tc.closure_dates(code, start="2026-01-01", end="2077-12-31"):
cal.addHoliday(ql.Date(d.day, d.month, d.year))
cal.adjust(ql.Date(29, 5, 2027), ql.ModifiedFollowing) # 28 May 2027Ask your AI assistant (Claude, ChatGPT, Cursor)
“Build the semi-annual schedule of a 10-year swap starting 30 November 2026 on New York and London days, modified following, end-of-month.”
Questions
Following, modified following, preceding, modified preceding and unadjusted; monthly to multi-year frequencies; short and long stubs at the front or back; the end-of-month rule; payment lags; and fixing dates with a lag on a separate calendar. Schedules can span up to 60 years.