Trading Calendar

You can use function getMarketCalendar(marketName, [startDate], [endDate]) to get trading days of the corresponding exchange in the date range determined by startDate and endDate.

To check the trading days of New York Stock Exchange (XNYS) between 2022.1.1 and 2022.1.10:

getMarketCalendar("XNYS",2022.01.01, 2022.01.10)

// output
[2022.01.03,2022.01.04,2022.01.05,2022.01.06,2022.01.07,2022.01.10]

To shift a trading day forward or backward, you can use function temporalAdd(date, duration, exchangeId).

Take XNYS for example, we add two trading days to the dates between 2023.1.1 and 2023.1.6:

dates=[2023.01.01, 2023.01.02, 2023.01.03, 2023.01.04, 2023.01.05, 2023.01.06]
temporalAdd(dates,2,"XNYS")

// output
[2023.01.04,2023.01.04,2023.01.05,2023.01.06,2023.01.09,2023.01.10]

Or you can also use the following script if you use version 2.00.11.1 or higher. For detailed usage of trading calendar as DURATION type, refer to section "Use Trading Days as DURATION Type".

dates=[2023.01.01, 2023.01.02, 2023.01.03, 2023.01.04, 2023.01.05, 2023.01.06]
temporalAdd(dates, 2XNYS)

//output
[2023.01.04,2023.01.04,2023.01.05,2023.01.06,2023.01.09,2023.01.10]

You can get the closest trading day of a certain day with function transFreq(X,rule).

For example, specify parameter rule as XNYS. We can get the closest trading days of each date between 2023.1.1 and 2023.1.6:

dates=[2023.01.01, 2023.01.02, 2023.01.03, 2023.01.04, 2023.01.05, 2023.01.06]
dates.transFreq("XNYS")

// output
[2022.12.30,2022.12.30,2023.01.03,2023.01.04,2023.01.05,2023.01.06]

You can choose functions asFreq(X,rule) or resample(X,rule,func) to sample data on trading days. The only difference of the two lies in whether data can be aggregated.

Function asFreq(X,rule) will return the result by trading days. If there are multiple records in the same trading day, only the first value will be taken. If there is no data in a trading day, it will be filled with NULL.

The following example obtains the stock prices of XNYS in trading days from 2022.12.30 to 2023.1.6:

timestampv = [2022.12.30T23:00:00.000,2023.01.01T00:00:00.000,2023.01.03T00:10:00.000,2023.01.03T00:20:00.000,2023.01.04T00:20:00.000,2023.01.04T00:30:00.000,2023.01.06T00:40:00.000]
close = [100.10, 100.10, 100.10, 78.89, 88.99, 88.67, 78.78]
s=indexedSeries(timestampv, close)
s.asFreq("XNYS")

// output
           #0                 
           ------
2022.12.30|100.10
2023.01.03|100.10
2023.01.04|88.99 
2023.01.05|                   
2023.01.06|78.78

Function resample(X,rule,func) will return the aggregated result of data sampled by trading days.

In the following example, we obtain the closing prices of XNYS stocks in trading days from 2022.12.30 to 2023.1.6:

timestampv = [2022.12.30T23:00:00.000,2023.01.01T00:00:00.000,2023.01.03T00:10:00.000,2023.01.03T00:20:00.000,2023.01.04T00:20:00.000,2023.01.04T00:30:00.000,2023.01.06T00:40:00.000]
close = [100.10, 100.10, 100.10, 78.89, 88.99, 88.67, 78.78]
s=indexedSeries(timestampv, close)
s.resample("XNYS", last)

// output
           #0                 
           ------
2022.12.30|100.10
2023.01.03|78.89
2023.01.04|88.67 
2023.01.05|                   
2023.01.06|78.78

Starting from version 2.00.11.1, trading calendar, specified as "integers + identifiers", can be used as DURATION data.

Suppose there is an exchange named "XDDB", function addMarketHoliday(marketName, holiday) can be used to add a new XDDB calendar. A XDDB.csv file will be added to the /marketHoliday/ directory. Weekends are recognized as holidays in DolphinDB by default, therefore, only weekday holidays need to be filled in the file.

Once a new trading calendar has been generated, functions such as getMarketCalendar can be used directly based on the new calendar:

//set 2023.01.03 (Tue.) and 2023.01.04 (Wed.) as holidays
holiday = 2023.01.03 2023.01.04  
//user login
login(`admin,`123456)
//generate a trading calendar
addMarketHoliday("XDDB",holiday)

//get the trading days of the new calendar in a date range
getMarketCalendar("XDDB",2023.01.01, 2023.01.10)
//output
[2023.01.02,2023.01.05,2023.01.06,2023.01.09,2023.01.10]

temporalAdd(2023.01.01,2,"XDDB")
//output
2023.01.05

Note: The newly added trading calendar is only valid on the current node. Execute function addMarketHoliday on other nodes for it to take effect on those nodes.

If you want to update the existing calendar of XDDB exchange, function updateMarketHoliday(marketName, holiday) can be used to reset the holidays.

Note: The file will be overwritten. The original holidays will be replaced with the holidays specified by this function.

The following example resets the dates 2023.03.07 and 2023.03.08 as holidays for the XDDB calendar. Check the next trading day after 2022.01.01 with function temporalAdd:

//set 2023.03.07 (Tue.) and 2023.03.08 (Wed.) as holiday
updateMarketHoliday("XDDB",2023.03.07 2023.03.08)

//the original holidays 2023.01.03 and 2023.01.04 are no longer holidays
getMarketCalendar("XDDB",2023.01.01, 2023.01.10)
//output
[2023.01.02,2023.01.03,2023.01.04,2023.01.05,2023.01.06,2023.01.09,2023.01.10]

//As holidays, 2023.03.07 and 2023.03.08 are not included in the trading calendar
getMarketCalendar("XDDB",2023.03.01, 2023.03.10)
//output
[2023.03.01,2023.03.02,2023.03.03,2023.03.06,2023.03.09,2023.03.10]