Andrea Carriero, Francesco Corsello, and Massimiliano Marcellino, "The
Global Component of Inflation Volatility", Journal of Applied
Econometrics, Vol. 37, No. 4, 2022, pp. 700-721.

All data and code are stored in the file ccm-files.zip. ASCII files
are in DOS format.

#####  DATA  #####

The data are taken from the Organisation for Economic Cooperation and
Development (OECD).  The data are downloadable from the OCSE Statistics
portal (OECD.stat: https://stats.oecd.org/), in the Consumer price
indices (CPIs) section. 

The file CCM_InflaHeadline.csv reports the year-on-year changes
(expressed in percentage points) of the overall (all items) CPI
Indexes for 20 countries (reported below) and 240 quarterly
observations, from 1960Q1 to 2019Q4. The first column of the file
contains the dates.

The file CCM_InflaCore.csv reports the year-on-year changes (expressed
in percentage points) of the core (all items non-food non-energy) CPI
Indexes for 20 countries (reported below) and 164 quarterly
observations, from 1979Q1 to 2019Q4. The first column of the file
contains the dates.

Both csv files are contained in the Data directory.

The dataset contains time series for the following 20 countries (in
alphabetical order): Australia, Austria, Belgium, Canada, Finland,
France, Germany, Greece, Italy, Japan, Luxembourg Netherlands, New
Zealand, Norway, Portugal, Spain, Sweden, Switzerland, United Kingdom,
United States


#####  CODES  #####

Programs are written in Matlab.

The Codes folder contains the main files to estimate the MAIARSV
model, produce the decompositions and the main graphical outputs of
the paper for the benchmark specification using data on headline
inflation. The main scripts are:

-- RunPosterior_MAIARSV_inflation.m is the primary one that runs the
   MCMC sampler to obtain the Posterior distribution of the MAIARSV
   specification using the inflation data.

-- Make_MAIARSV_decompDistr.m produces the distribution of the
   decompositions of levels and volatilities described in section 2.

-- plot_fact_infla.m produces the pictures reporting observable data
   along with the global factor (with posterior bands)

-- plot_infla_decomp_MAIARSV.m produces all the pictures regarding the
   decompositions proposed in the paper and computed by the script
   Make_MAIARSV_decompDistr.m

-- plot_analysis_persistence_inflation.m calculates and plots the
   largest roots (according to Stock (1991) methodology) of observable
   inflation data, and their common and individual components. 

The subfolder functions include several important MATLAB functions
that are used throughout the scripts. These include, among others:

-- build_prior_std_MAIAR_SV.m creates the priors used in the MAIARSV
   model and described in section 3.

-- RW_metro_step_b0.m runs the Metropolis Step to draw B0

-- logLikKernel_MAI_VAR_SV.m compute the log-likelihood used in the
   Metropolis Step to draw B0

-- FFBS_kfilter.m and FFBS_sampling.m run the Forward Filter Backward
   Sampling algorithm by Carter and Kohn (1994)

-- log_chisquared1_approx_states.m computes the matrix containing the
   indexes of components of the mixture of normal distributions
   approximating the log_chi_squared_1 as in Kim, Shephard, Chib (1998)
   and Omori, Chib, Shephard and Nakajima (2007)

-- CTA.m performs a draw from the conditional posterior of a VAR
   conditional on mean coefficients by using the triangular algorithm