Skip to content
Replication of FRBNY DSGE model in IRIS
Branch: master
Clone or download
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
README.md Update README.md Apr 14, 2017
change_parameters_and_sstates.m added nonlinear Kimball aggregator for prices May 23, 2018
check_loglin.m
compare_model_and_data.m fixed #2 Aug 28, 2017
estimate_params.m
filter_hist_data.m added nonlinear Kimball aggregator for prices May 23, 2018
fisher_information_matrix.m latest commit Apr 12, 2017
forecasts_with_judgment.m latest commit Apr 12, 2017
frbny.model
get_info_about_model.m latest commit Apr 12, 2017
helper_plot_acf.m
helper_plot_xsf.m added nonlinear Kimball aggregator for prices May 23, 2018
information.m added nonlinear Kimball aggregator for prices May 23, 2018
knitronlp.m benchmarking fmincon vs Knitro active set SQP algorithm Nov 9, 2017
knitronlp.opt added nonlinear Kimball aggregator for prices May 23, 2018
know_all_about_solution.m
more_on_kalman_filter.m latest commit Apr 12, 2017
more_on_poster_simulator.m latest commit Apr 12, 2017
ois_blp.mat
priors.m benchmarking fmincon vs Knitro active set SQP algorithm Nov 9, 2017
read_data.m
read_data.mat
read_linear_model.m added nonlinear Kimball aggregator for prices May 23, 2018
read_me_first.m replicated zero lower bound Feb 18, 2017
read_nonlin_model.m added nonlinear Kimball aggregator for prices May 23, 2018
resample_from_model.m
simulate_complex_shocks.m latest commit Apr 12, 2017
simulate_disinflation.m latest commit Apr 12, 2017
simulate_simple_shock.m
simulate_simple_shocks.m added nonlinear Kimball aggregator for prices May 23, 2018
solve_sigw.m added tutorial Nov 30, 2016
zeta_sp.m

README.md

FRBNY-IRIS

FRBNY model in IRIS

The IRIS toolbox is available at https://github.com/IRIS-Solutions-Team/IRIS-Toolbox.git

frbny.model is the model file for IRIS toolbox. It initializes all parameters, lists endogenous variables and exogenous shocks in FRBNY model's linear and nonlinear equations.

estimate_params.m - main program that reads the model file (linear or nonlinear version), runs estimation and does MH sampling. Optimization in IRIS is done using fmincon which is about 30 times faster than using csminwel with gensys.

filter_hist_data.m - shock decomposition and historical simulations

estimate_params.mat - contains the parameter struct est with the max posterior mode

priors.m - contains the priors and lower and upper bounds on parameters

Unconstrained minimization algorithm such as csminwel requires rescaling of constrained parameters. The problem is that we don't know if any of the constraints will become binding at the mode. For example, parameter rho_sigw becomes binding at the upper boundary 0.99 that was imposed in original v990 code (as evidenced from output from IRIS optimizer fmincon). That caused the csminwel to slow down to a crawl since the unscaled parameter tried to go to infinity. I therefore had to increase this upper limit to 0.999 (since the mode is at 0.9945). As a result the Julia csminwel code which was running several days, now runs in 4 hours after binding constraint on rho_sigw was removed. We can avoid this problem in IRIS by using constrained minimization method such as fmincon (with active-set algorithm), which can estimate the model in 5 minutes without zero lower bound and in 15 with zero lower bound.

You can’t perform that action at this time.