Matlab Codes

DSGE: contains files to compute solutions to DSGE models

_ chomoreno.m: computes solutions to first order using the Forward method of Cho and Moreno. Under indeterminacy, the code directly computes the solutions using the method developed by Lubik and Schorfheide. If requested by the user, the latter solution method can be employed also under determinacy.

_ lombsuth.m: computes second order solutions using the Lombardo-Sutherland method. The first order solution is computed using the Forward method.

_ lubik.m: example code containing the New Keynesian model in Lubik and Schorfheide, JEDC (2003). This code uses chomoreno.m to obtain solutions under indeterminacy. The impulse responses correspond to the orthogonal solution, as defined in Lubik and Shorfheide (2003), and the Forward solution. The latter solution corresponds to the continuity solution as defined in Lubik and Shorfheide (2003).

_ lombardo_sutherland_RBC.m: example code containing the RBC model in Lombardo and Sutherland, JEDC (2007). This code uses lombsuth.m to obtain second order solutions. The model has to be rewritten in such a way as to get rid of all static and mixed variables. The code dyname_lombsuth.mod is a dynare code with the same model, which can be used to compare the results.

Vector AutoRegression: contains files to compute SVARs with recursive, long-run and proxy variable identification.

_srSVAR.m: computes impulse responses and variance decomposition of a recursive VAR (Monte Carlo bands).

_lrSVAR.m: computes impulse responses and variance decomposition of a SVAR with long run (Blanchard-Quah) restrictions (Monte Carlo bands).

_granger.m: computes Granger tests for the VAR.

_ivSVARwildbootstrap.m: computes impulse responses of a SVAR identified with a proxy variable (confidence bands are obtained with a wild bootstrap approach).

_ivSVARblockbootstrap.m: computes impulse responses of a SVAR identified with a proxy variable (confidence bands are obtained with a block bootstrap approach).

_girfSVAR.m: computes impulse responses and variance decomposition of a VAR with the generalized method of Pesaran and Shin (EL, 1998) (Monte Carlo bands).

_signrestSVAR.m: computes impulse responses and variance decomposition of a VAR identified with sign restrictions.

_main.m: runs the VAR estimations on US output and inflation. The data are in the excel file data.xmls. Stata do.files running some of the VARs are included for comparison.