Manfred M. Fischer, Niko Hauzenberger, Florian Huber, and Michael 
Pfarrhofer, "General Bayesian Time-Varying Parameter VARs for Modeling 
Government Bond Yields", Journal of Applied Econometrics, Vol. 38, No. 1,
2023, pp. 69-87.

All data files are contained in fhhp-data.zip. The .csv files are in DOS 
format.

We use monthly zero-coupon yields of US treasuries at yearly
maturities (1,3,5,7,10 and 15 years). These yields are publicly
available online at

https://www.federalreserve.gov/data/nominal-yield-curve.htm.

We provide the raw data (at a daily frequency) as a .csv file
("US_YLDs.csv"), where the first column contains the dates and the
columns labelled "SVENYxx" refer to the zero-coupon yields at a yearly
maturity. For the empirical application, we consider SVENY01, SVENY03,
SVENY05, SVENY07, SVENY10 and SVENY15 as our target variables after
averaging them to monthly frequency.

In addition, we use three observed factors as exogenous effect
modifiers (provided in separate .csv files):

1.) A binary recession indicator for the US, dated on a monthly basis
by the Business Cycle Dating Committee of the National Bureau of
Economic Research and downloaded from the FRED database of the Federal
Reserve Bank of St. Louis (available online via

https://fred.stlouisfed.org/series/USREC

and provided as "US_REC.csv").

2.) The National Financial Conditions Index (NFCI), maintained by the
Federal Reserve Bank of Chicago and downloaded from the FRED database
(available online via

https://fred.stlouisfed.org/series/NFCI

and provided as "US_NFCI.csv").

3.) The Risk-Free (RF) interest rate from the Fama-French Portfolios
and Factors database, downloaded from the web page of Kenneth R.
French (available online at

https://mba.tuck.dartmouth.edu/pages/faculty/ken.french

and provided as "US_FF.csv", where the column "RF" refers to the
Risk-Free rate).

We also provide a collected version of the data as an .rda-object
("fhhp_jae.rda"). This file contains two data frames. While "ylds"
stores the monthly zero-coupon yields for our target maturities, "exo"
collects the exogenous effect modifiers. The first column "TIME" of
the two data frames contains the dates. Both data sets cover the
periods from 1973:01 to 2019:12 (i.e., 564 observations).

R codes for estimating the models are available upon request.


Contact information:

Niko Hauzenberger (University of Salzburg)
Address: Moenchsberg 2A, 5020 Salzburg, Austria
Email: niko.hauzenberger [at] sbg.ac.at