Calendars & timezones
This page demonstrates how to configure the timezone and calendar features when initializing your environment. It also explains how to search and work with the schedule and calendar functions available within the framework.
This information would be valuable for any SigTech user wanting to build strategies which trade outside of the UTC timezone and trading calendar.
When initializing your environment with the
sig.init() command, calendar and timezone settings will be configured by default.import sigtech.framework as sig
import datetime as dtm
import pandas as pd
import numpy as np
sig.init()
A
TradingManager object is automatically created during environment initialization. These objects manage calendar, timezone and open/close time information for strategies. They also provide a collection of useful trading functionalities. Configuration parameters including the initials TM are referencing the TradingManager.Parameter | Default |
|---|---|
TM_CALENDAR | 'None' |
TM_TIMEZONE | 'Europe/London' |
TM_OPEN | datetime.time(7, 30, 0) |
TM_CLOSE | datetime.time(18, 0, 0) |
TM_DEFAULT_DATA_POINTS | [DataPoints.LONDON_1600, DataPoints.LAST] |
TM_INTRADAY_OUT_OF_HOURS | False |
Available timezones are included in the standard python timezone library,
pytz. To generate a list of the available options, you can run the following command:import pytz
pytz.all_timezones
Available calendars are listed in the Calendars folder of our Data browser: available calendars. The object name can be retrieved from the VIEW DETAILS page, under CODE EXAMPLE.
The most commonly used calendars are:
'BANGKOK': 'BANGKOKB CALENDAR',
'BUDAPEST': 'BUDAPESTB CALENDAR',
'HONG KONG': 'HONGKONGB CALENDAR',
'ISTANBUL': 'ISTANB CALENDAR',
'JOHANNESBURG': 'JOHANNESBB CALENDAR',
'LONDON': 'LONDONB CALENDAR',
'MEXICO CITY': 'MEXICOB CALENDAR',
'MOSCOW': 'MOSCOWB CALENDAR',
'MUMBAI': 'MUMBAIB CALENDAR',
'NEW YORK': 'FEDFUND CALENDAR',
'OSLO': 'OSLOB CALENDAR',
'PARIS': 'PARISB CALENDAR',
'SEOUL': 'SEOULB CALENDAR',
'SINGAPORE': 'SINGAPOREB CALENDAR',
'STOCKHOLM': 'STOCKHOLMB CALENDAR',
'SYDNEY': 'SYDNEYB CALENDAR',
'TARGET': 'TARGET CALENDAR',
'EUR': 'TARGET CALENDAR',
'TOKYO': 'TOKYOB CALENDAR',
'TORONTO': 'TORONTOB CALENDAR',
'WARSAW': 'WARSAWB CALENDAR',
'WELLINGTON': 'WELLINGTB CALENDAR',
'USD': 'FEDFUND CALENDAR',
'ZURICH': 'ZURICHB CALENDAR'
The Data Browser lists all available calendars, and provides sample code showing how to get them. For example, to get the A2X South Africa calendar:
calendar = sig.obj.get("A2XSE(T) CALENDAR")
# Description
print(calendar.description)
# List all holidays
print(calendar.holidays)
To identify relevant calendars for currency pairs, use
FXMarket.instance() to retrieve specific schedules.Input
Output
from sigtech.framework.infra.analytics.fx.fx_market import FXMarket
FXMarket.instance().schedule_stub('HKD', 'USD').holidays
'HKD CALENDAR,WMCO CALENDAR'
For futures based strategies, you can retrieve exchange calendars directly from the futures group with the following code.
Input
Output
start_date = dtm.date(2010, 1, 4)
strategy = sig.RollingFutureStrategy(
currency='USD',
start_date=start_date,
contract_code='GC',
contract_sector='COMDTY',
rolling_rule='front',
front_offset='-5:-4',
total_return=False,
include_trading_costs=False
)
strategy.contract_group.holidays
'CMECOMF(T) CALENDAR'
You can then retrieve their relevant holiday calendars.
Input
Output
sig.obj.get('COU21 COMDTY').group().holidays
'ICE_EU_IPE(T) CALENDAR'
For equities based strategies, you can retrieve exchange calendars with the following command.
holiday_calendar = sig.obj.get('1000045.SINGLE_STOCK.TRADABLE').exchange().holidays
For business days, you will need to fetch the
holiday_calendar object and then call business_days, specifying a start and end date.business_days = sig.obj.get(holiday_calendar).business_days(
start=dtm.date(2020, 1, 1),
end=dtm.date(2022, 3, 15)
)
OTC instruments contain a
.holiday method which retrieves the relevant calendars without the need for calling sig.obj.get.Input
Output
irs = sig.InterestRateSwap(currency='USD',
trade_date=dtm.date(2017, 4, 14),
start_date=dtm.date(2017, 5, 3),
tenor='5Y')
irs.history();
irs.holidays
'FEDFUND, LONDONB CALENDAR'
You can reconfigure the calendar and timezone values during initialization. However, if you have already initialized your environment the
sig.de_init() call will allow you to reset the values. The following code block will change the defaults.# Destroy your environment
sig.de_init()
# Reconfigure your environment
env = sig.init()
sig.config.set('TM_TIMEZONE', 'Asia/Tokyo')
sig.config.set('TM_CALENDAR', 'HKD CALENDAR')
This code block can be used if initializing your environment for the first time by simply removing the
sig.de_init() command.You can check these changes have been successful by running the following commands:
Input
Output
env[sig.config.TM_TIMEZONE]
'Asia/Tokyo'
Input
Output
env[sig.config.TM_CALENDAR]
'HKD CALENDAR'
You can add or adjust business day conventions, and retrieve month-end and IMM dates using
sig.calendar.Input
Output
# adding business days
sig.calendar.date_increment(
dtm.date(2021, 12, 1),
'1W',
calendar='HKD CALENDAR',
bdc='FOLLOWING'
)
datetime.date(2021, 12, 8)
Input
Output
# month-end
sig.calendar.get_month_end_dates(
start=dtm.date(2021, 9, 1),
end=dtm.date(2022, 1, 3),
calendar_names='WMCO CALENDAR'
)
[datetime.date(2021, 9, 30),
datetime.date(2021, 10, 29),
datetime.date(2021, 11, 30),
datetime.date(2021, 12, 31)]
Input
Output
# IMM dates
sig.calendar.imm_date(ref_date=dtm.date.today(), currency='USD')
datetime.date(2022, 6, 15)
SigTech supports the majority of common rebalancing schedules.
Using a
BasketStrategy as an example, the AVAILABLE_REBALANCE_FREQS object can help you to locate the correct string to supply for rebalancing.Input
Output
freqs = sig.BasketStrategy.AVAILABLE_REBALANCE_FREQS
frequencies = [f for f in dir(freqs) if not f.startswith('_')]
frequency_strings = sorted([f for f in frequencies if f == f.upper()])
frequency_strings
['END_OF_MONTH', 'IMM', 'START_OF_MONTH', 'YEARLY']
You can use the values outputted above to determine the correct input string:
Input
Output
print(freqs.END_OF_MONTH)
print(freqs.IMM)
print(freqs.START_OF_MONTH)
print(freqs.YEARLY)
EOM
3M_IMM
SOM
YEARLY
Input
Output
frequency_methods = sorted([f for f in frequencies if f == f.lower()])
frequency_methods
['business_days',
'day_of_month',
'fridays',
'mondays',
'months',
'thursdays',
'tuesdays',
'wednesdays',
'weeks']
The values outputted above are methods that take an input and return the correct string format. You can use the
thursdays method is an example:Input
Output
freqs.thursdays?
Signature: freqs.thursdays(n=1)
Docstring:
Rebalance every ``n`` weeks on Thursdays.
:param n: Number of weeks, 1 by default.
:return: Rebalance string value.
File: /opt/conda/envs/sig-env/lib/python3.7/site-packages/sigtech/framework/schedules/schedule.py
Type: method
The input
n determines how often, in terms of weeks, to rebalance. For instance, freqs.thursdays(2) returns the string to input to rebalance a strategy every second Thursday.The following cell block runs some of these methods and outputs the strings necessary for the
rebalancing_frequency.Input
Output
print(freqs.business_days(2))
print(freqs.day_of_month(2))
print(freqs.fridays(2))
print(freqs.thursdays(2))
print(freqs.weeks(2))
2BD
2DOM
2W-FRI
2W-THU
2W
SigTech includes a range of rebalancing schedules; EOD, 1W, 2W, 1M, SOM, EOD, 2BD, 2DOM, 2W-FRI, 2W-THU, 2W.
The rebalancing schedules listed above are not exhaustive. The Rebalancing frequencies section of this page provides the necessary code to list all of the available options.
If your desired rebalancing frequency is unavailable, a custom schedule can be created from the
SchedulePeriodic class.Python
Output
from sigtech.framework.schedules import SchedulePeriodic
end_date = dtm.date(2021, 5, 14)
frequency = '1W-Fri'
# Custom schedule for 3rd Friday Quarterly Rebalance
custom_schedule = SchedulePeriodic(
start_date=start_date,
end_date=end_date,
holidays='NYSE(T) CALENDAR',
frequency=frequency,
).all_data_dates()
custom_schedule = [x for x in custom_schedule
if x.month in [3, 6, 9, 12]] # Quarterly rebalancing
custom_schedule = [x for x in custom_schedule
if x.day >= 15 and x.day <= 21] # third Friday
custom_schedule[-5:]
[datetime.date(2020, 3, 20),
datetime.date(2020, 6, 19),
datetime.date(2020, 9, 18),
datetime.date(2020, 12, 18),
datetime.date(2021, 3, 19)]
Last modified 1yr ago