rates: LIBOR rates from 2004-01-01 to 2015-08-03

Description Format Details Source See Also Examples


This data frame is created by build_rates in creditr package to calculate the CDS pricing. It covers three currencies: USD, EUR, and JPY. The interest rates date from 2004-01-01 to 2014-08-23. Rates on holidays and weekends are available as well as business days.


A data frame with 194378 observations on the following 4 variables.


The source of the interest rates in rates.RData is from https://www.markit.com/ and http://research.stlouisfed.org/fred2/. When a user is calculating CDS using the CDS package, the package calls get_rates to get the needed interest rates; get_rates then calls the rates.RData for these interest rates. If a date is unavailable in rates.RData, then the package calls other functions to get the needed interest rates from the internet. The rates.RData is created and stored in the package for the users' convenience: getting interest rates from the Internet may fail due to the internet connection problem and may be very slow. Also, the user can build its own updated local rates.RData by using build_rates. For more explanation on the usage of build_rates, please see See Also.

Also, please notice that in the rates.RData, the rate is not the interest rate on the date listed in the same row as the rate. For example, in the first row, the rate is 0.001550; the date in that row is 2014-08-21. But this does not mean that on 2014-08-21, the interest rate is 0.001550; instead, 0.001550 is the interest rate on 2014-08-20. This regulation of using the interest rate on the previous business day of CDS trading date is set by ISDA Standard Model. The Model says that, if a trader buy a CDS on 2014-08-21, then when calculating the pricing of the CDS, she should use the interest rate of 2014-08-20, which is 0.001550. This may be confusing for people who are yet unfamiliar with CDS. We design rates.RData in this way because it is easier for other functions in the CDS package to use this data frame for calculation.

rates.RData covers holidays, weekend and business days. As is set by ISDA Standard Model and introduced above, we use the previous business day's interest rate for CDS pricing on a certain trading date. Therefore, if the user is buying a CDS on Saturday, she should use the interest rate of last Friday; if she is buying a CDS on Sunday or Monday, she should still use the interest rate of last Friday, because last Friday is the previous business day of the CDS trading date. When it comes to holidays, we still choose the previous business day for interest rate. For example, if a trader is buying a CDS on 2014-07-05, then she should use the interest rate of 2014-07-03, because 2014-07-04 is a national holiday and 2014-07-03 is the previous business day of trading date.

Also, please notice that in rates.RData, a currency's type of expiries generally stays the same along the time, but not always. For example, we check the expiry type of USD in rates.RData: for 2004-01-01, there are two types of expirty: 1M, 3Y; for 2006-01-01, there are five types of expiry: 1M, 2M, 3M, 6M, 1Y; for 2007-01-01, there are 18 types of expiry; for 2014-01-01, there are 19 types of expiry. Therefore, for a trader, she should always keep in mind to update her knowledge of the expiry types of her trading currency. For different currencies, the types of expiries are often different.

Finally, please notice that some of the data are missing for a certain expiry of a currency in a short time. For example, some dates in rates.RData do not have a expiry of 3Y for USD. This is not likely to be caused by data error, since all these data are got from Markit and FRED. Users, however, should be aware of that some data seem "missing".


https://www.markit.com/ http://research.stlouisfed.org/fred2/

See Also

download_FRED download_markit build_rates



## for JPY rates:
rates[rates$currency == "JPY",]

## for rates on a specific date, of a specific currency:
rates[rates$currency == "USD" & rates$date == "2005-10-01",]

creditr documentation built on May 29, 2017, 8:46 p.m.