creditr (version 0.6.1)

build_rates: Build a data frame containing interest rates for CDS pricing

Description

build_rates can create a data frame of interest rates from a start date to an end date, which are specified by the user. The interest rates will be later used in CDS pricing. build_rates also builds the rates.RData in the package. build_rates mainly consists download_markit and download_FRED. The two sources are Markit website and FRED website.

Usage

build_rates(start.date, end.date)

Arguments

start.date

is the start date of the data frame; it is the earliest CDS pricing date that the user concerns.

end.date

is the end date of the data frame; it is the lastest CDS pricing date that the user concerns.

Value

a data frame that contains the CDS pricing date, the currency, the interest rate expiry and the interest rate.

Details

since Markit wesite has the interest rates back to th 1990s, download_markit is responsible for building up all the USD interest rate data frame; for EUR and JPY, markit can only get from 2005-01-05 to now. But the biggest advantage of using download_markit is that, since Markit website only lists the rate expiries that are actually used for CDS pricing, download_markit can get the exact type of expiries of rates needed to price CDS. Also, it has expiry over 1Y to 30Y. In contrast, FRED is only complementary to markit data. Since download_markit can get any rates for USD, we don't use FRED for USD. Then we want to get data for EUR and JPY before 2005-01-04 (the limit of markit). FRED has almost all data for any date, which is its biggest advantage. But its biggest disadvantage is that it doesn't know which expiry type is suitable for which time, since FRED website has expiry of all types below a year; also it doesn't have expiry over 1Y.So we hardcoded the dates below, to combine markit and FRED in the most suitable way.

Another note is that, the rates on both Markit website and FRED website have not been adjusted to the previous business day. In other words, the rates from both website is the exact rate on that day, rather than on the previous business day. But download_markit and download_FRED have adjusted the days already for the convenience of CDS pricing, so we don't have to worry here.

References

http://www.markit.com/ http://research.stlouisfed.org/fred2/categories/33003/downloaddata

See Also

download_markit download_FRED rates

Examples

## Not run: 
# ## Running this example will take more than two hours.
# 
# build_rates(start.date = as.Date("2004-01-01"),
#             end.date = as.Date("2014-08-23"))## End(Not run)