Benjamin Nelson, Gabor Pinter, and Konstantinos Theodoridis, "Do 
Contractionary Monetary Policy Shocks Expand Shadow Banking?", Journal 
of Applied Econometrics, Vol. 33, No. 2, 2018, pp. 198-211.

The data used in the article are from Flow of Funds website 
(http://www.federalreserve.gov/releases/Z1/about.htm). This data source 
and the rest of our macroeconomic data are publicly available. The Excel 
file "dataset_NPT_JAE.xls" contains the data together with exact 
definitions and sources. The matlab file NPT_baseline.m is used to 
generate the baseline IRF figure 3 of the paper as well as figure 5 
(historical counterfactual).

variable name:	definition and source

y	log real GDP (original source: BEA)

cpi	log of Personal Consumption Expenditures (original source: BEA)

r	Effective Federal Funds Rate (original source: BEA)

com	log of commercial bank (credit unions + depository institutions)
        assets [deflated by cpi] (original source: flow of funds)

shad	log of shadow bank (ABS-issuers + funding corporations +
        Mortgage pools + GSEs) [deflated by cpi]  (original source:
        flow of funds)

romer	narrative measure of Romer-Romer (AER, 2004), updated by O. Coibion

gilchrist  credit spread measure of Gilchrist-Zakrajsek ("Credit Spreads
           and Business Cycle Fluctuations", AER, 2012)

mortgage_spread	spread between the 30-year fixed rate mortgage average
                (MOLRTGAGE30US from BEA) and the 10-year constant maturity
                treasury rate (DGS10 from BEA)

Shiller	real home price index from Robert Shiller's website
        (http://www.econ.yale.edu/~shiller/)

term	spread between the 10-year constant maturity treasury rate
        (DGS10 from BEA) and the 3-month Treasury bill rate (TB3MS from BEA)

The NPT_baseline.m function reproduces the SVAR results in the text.

The zip files DynareFiles.zip and npt-functions.zip contain the matlab 
functions that implement the Bayesian MDE and DSGE simulations in the 
paper. These functions use Dynare 4.4.3.

The exe_dynare_estimation.m function implements the Bayesian MDE 
estimation of the model The exe_dsge_simulations.m function produces the 
simulations of the DSGE model.

Please address any questions to:

Konstantinos Theodoridis
Bank of England
EC2R 8AH
Threadneedle Street
London, UK
Konstantinos.Theodoridis [AT] bankofengland.co.uk