Andrea Carriero, Todd E. Clark, and Massimiliano Marcellino, "No
Arbitrage Priors, Drifting Volatilities, and the Term Structure of
Interest Rates", Journal of Applied Econometrics, Vol. 36, No. 5,
2021, pp. 495-516.

This file details data and computer programs made available to estimate 
the JSZ-VAR-CSV model developed in the paper. All files are zipped in 
the file ccm-files.zip. Most files are ASCII files in DOS format.

The paper's results use data on zero-coupon yields, at monthly
frequency, for maturities 1 and 3 months and 1, 2, 3, 4, 5, 7, and 10
years. We obtained the yields from maturities of 1 month through 5
years from the US Treasuries Daily and Monthly database of the Center
for Research in Security Prices (CRSP), the University of Chicago
Booth School of Business. We took the 7- and 10- year yields from the
Gurkaynak, Sack, and Wright (2007) data published by the Federal
Reserve Board of Governors.  Licensing restrictions prevent us from
providing the CRSP data we used.

As a substitute, for the purpose of allowing interested researchers to
estimate our model with our code provided, we include in this archive
yields for the full set of maturities indicated above, from Liu and
Wu, "Reconstructing the Yield Curve," Journal of Financial Economics.
Updates of the data are available from the website of Prof. Wu, at
this link: 
https://sites.google.com/view/jingcynthiawu/yield-data?authuser=0


**************
File contents:  
**************

Programs are written in Matlab.

The root folder contains the data file LiuWudata.xlsx in Excel format.
The file includes 9 monthly yields (1mo, 3mo, 1yr, 2yr, 3yr, 4yr, 5yr,
7yr, & 10yr) from years 1985 to 2018, with January of year xxxx
labeled as date xxxx.08 and December of the same year labeled as date
(xxxx+1).00.  Note that we write an interest rate of, for example, 2
percent as 0.02 (i.e., the data in our file are 0.01 times the data as
we obtained them from the source Liu and Wu file).

The Matlab file V2020_main_JESubmission_JSZCSVVAR_ONLY.m is the
primary one for estimating the JSZ-VAR-CSV specification.  In this
file we build the BVAR, priors, and posteriors, calling various other
Matlab files to perform specific functions.  These include, among
others:

-- create_prior.m creates the priors used in the JSZ-CSV-VAR model.

-- mlikelihood.m and mlikelihood_condTight build the maximum
   likelihood estimates for the JSZ-CSV-VAR model.

-- Moments.m generates the moments used in the prior likelihood and
   posterior distribution.

-- posterior.m and posterior_hessian build the posterior distributions
   for JSZ-CSV-VAR.

-- csminSIMS contains relevant .m files for creating minimizing functions.

-- distrib contains functions used in some distributions computations.

-- gibbs contains files for Gibbs sampling.

-- jsz_library contains files from Joslin, Singleton, and Zhu used to
   build the affine term structure model-based prior

-- Results3 is a directory to which results are written

-- util has other necessary functions.


Notes regarding pieces of V2020_main_JESubmission_JSZCSVVAR_ONLY.m:

The data file is named LiuWudata.xlsx (Line 116).

Selecting model parameters are largely at the beginning of the code.

Model 4 is the main model, JSZ-VAR-CSV (Line 106 : ModelSeq=4;).

Next, select Tstart = 120 and Tlast = 408 - 12 + 1 where each value is
one month (Lines 148-149).  Allowing for the Zero-Lower-Bound, this
takes a couple months to output.  If doing, run two or three MATLAB to
fasten up pace.

Figures 1, 2, 3 are produced via Verbose = 1 (Line 91).

Table 1 is produced by manually setting lam_value to what you want. If
unspecified, integrated out. Table 1a is produced by doing this
ModelSeq = 4 and then, Tlast = 409 - 12 + 1 and then, Tstart = Tlast.

NOTE: The program files use different greek letters or letter
spellings than in the paper itself.


                code | paper
             -------------------
                     |
                teta |  theta Equation (5)
                     |
                vol  |  lambda
                     |
                psi  |  PHI VAR Coefficeints in Equation (8)
                     |
              lambda |  gamma Tightness Parameter
                     |
               sigma |  V Constant Volatility in Equation (9)
                     |
            voladraw |  LAMBDA
                     |
               loght |  log(lambda_t) Equation (10)
                     |


Contact information:  a.carriero [AT] qmul.ac.uk