diff --git a/doc/dynare.texi b/doc/dynare.texi index 3bbaf9c6578daf902a21787e364916e5ddb0381d..7ecf288d836b51ed2dacd145f5d415333e6d73bc 100644 --- a/doc/dynare.texi +++ b/doc/dynare.texi @@ -7082,7 +7082,7 @@ cannot be less than the number of constrained periods. Number of simulations. Default: @code{5000}. @item conf_sig = @var{DOUBLE} -Level of significance for confidence interval. Default: @code{0.80} +Level of significance for confidence interval. Default: @code{0.90} @end table diff --git a/matlab/bvar_forecast.m b/matlab/bvar_forecast.m index 1b88e0ae61552c32fad4d0c79ee976a3be268f89..9b22ff679ee60333320245cbb4a327ea6cc082ac 100644 --- a/matlab/bvar_forecast.m +++ b/matlab/bvar_forecast.m @@ -103,7 +103,7 @@ end % Plot graphs sims_no_shock_mean = mean(sims_no_shock, 3); -sort_idx = round((0.5 + [-options_.conf_sig, options_.conf_sig, 0]/2) * options_.bvar_replic); +sort_idx = round((0.5 + [-options_.bvar.conf_sig, options_.bvar.conf_sig, 0]/2) * options_.bvar_replic); sims_no_shock_sort = sort(sims_no_shock, 3); sims_no_shock_down_conf = sims_no_shock_sort(:, :, sort_idx(1)); diff --git a/matlab/bvar_irf.m b/matlab/bvar_irf.m index a2e124f3a3717ec6ded3c0f0ccc56eeacad8228b..17228b5c433b3298e4dea1b9b8fdee825bfb86b9 100644 --- a/matlab/bvar_irf.m +++ b/matlab/bvar_irf.m @@ -102,7 +102,7 @@ posterior_mean_irfs = mean(sampled_irfs,4); posterior_variance_irfs = var(sampled_irfs, 1, 4); sorted_irfs = sort(sampled_irfs,4); -sort_idx = round((0.5 + [-options_.conf_sig, options_.conf_sig, .0]/2) * options_.bvar_replic); +sort_idx = round((0.5 + [-options_.bvar.conf_sig, options_.bvar.conf_sig, .0]/2) * options_.bvar_replic); posterior_down_conf_irfs = sorted_irfs(:,:,:,sort_idx(1)); posterior_up_conf_irfs = sorted_irfs(:,:,:,sort_idx(2)); diff --git a/matlab/forcst.m b/matlab/forcst.m index 2684df8ee86b805f9d0a1643d8eb855748ed4b9b..0cc8dda2d245b6bb63e65e2e43f7b9b7bc13aeb4 100644 --- a/matlab/forcst.m +++ b/matlab/forcst.m @@ -78,7 +78,7 @@ for i=1:horizon sigma_y = sigma_y+sigma_u; end -fact = norminv((1-options_.conf_sig)/2,0,1); +fact = norminv((1-options_.forecasts.conf_sig)/2,0,1); int_width = zeros(horizon,M_.endo_nbr); for i=1:nvar diff --git a/matlab/forecast_graphs.m b/matlab/forecast_graphs.m index 5dab918116823621bbbf5b00dbe8546cec0b2855..ea1932e856bcc5a8a7bcbbd8e9f4d96471705058 100644 --- a/matlab/forecast_graphs.m +++ b/matlab/forecast_graphs.m @@ -74,7 +74,7 @@ for j= 1:nvar fprintf(fidTeX,'\\centering \n'); fprintf(fidTeX,'\\includegraphics[scale=0.5]{%s/graphs/forcst%d}\n',dname,n_fig); fprintf(fidTeX,'\\label{Fig:forcst:%d}\n',n_fig); - fprintf(fidTeX,'\\caption{Mean forecasts and %2.0f%% confidence intervals}\n',options_.conf_sig*100); + fprintf(fidTeX,'\\caption{Mean forecasts and %2.0f%% confidence intervals}\n',options_.forecasts.conf_sig*100); fprintf(fidTeX,'\\end{figure}\n'); fprintf(fidTeX,' \n'); end @@ -110,7 +110,7 @@ if m > 1 fprintf(fidTeX,'\\centering \n'); fprintf(fidTeX,'\\includegraphics[scale=0.5]{%s/graphs/forcst%d}\n',dname,n_fig); fprintf(fidTeX,'\\label{Fig:forcst:%d}\n',n_fig); - fprintf(fidTeX,'\\caption{Mean forecasts and %2.0f\\%% confidence intervals}\n',options_.conf_sig*100); + fprintf(fidTeX,'\\caption{Mean forecasts and %2.0f\\%% confidence intervals}\n',options_.forecasts.conf_sig*100); fprintf(fidTeX,'\\end{figure}\n'); fprintf(fidTeX,' \n'); end diff --git a/matlab/global_initialization.m b/matlab/global_initialization.m index a5844b9968be22fa902f6c07ac450c9dffe5b9e5..e5a9a8e44c363289489b22d00654b983640889a1 100644 --- a/matlab/global_initialization.m +++ b/matlab/global_initialization.m @@ -109,6 +109,7 @@ options_.bvar_prior_mu = 2; options_.bvar_prior_omega = 1; options_.bvar_prior_flat = 0; options_.bvar_prior_train = 0; +options_.bvar.conf_sig = 0.6; % Initialize the field that will contain the optimization algorigthm's options declared in the % estimation command (if anny). @@ -305,6 +306,8 @@ options_.prior_draws = 10000; options_.sampling_draws = 500; options_.forecast = 0; +options_.forecasts.conf_sig = 0.9; +options_.conditional_forecast.conf_sig = 0.9; % Model options_.linear = 0; @@ -513,7 +516,7 @@ options_.estimation.moments_posterior_density.gridpoints = 2^9; options_.estimation.moments_posterior_density.bandwidth = 0; % Rule of thumb optimal bandwidth parameter. options_.estimation.moments_posterior_density.kernel_function = 'gaussian'; % Gaussian kernel for Fast Fourrier Transform approximaton. % Misc -options_.conf_sig = 0.6; +% options_.conf_sig = 0.6; oo_.exo_simul = []; oo_.endo_simul = []; ys0_ = []; diff --git a/matlab/imcforecast.m b/matlab/imcforecast.m index b9ffeac4e33f80d60cac14a512620d9ed9df5d5c..e310a642bde4fd1cf95073bd9de59bb10bda7960 100644 --- a/matlab/imcforecast.m +++ b/matlab/imcforecast.m @@ -63,8 +63,8 @@ if ~isfield(options_cond_fcst,'periods') || isempty(options_cond_fcst.periods) options_cond_fcst.periods = 40; end -if ~isfield(options_cond_fcst,'conf_sig') || isempty(options_cond_fcst.conf_sig) - options_cond_fcst.conf_sig = .8; +if ~isfield(options_cond_fcst,'conditional_forecast') || ~isfield(options_cond_fcst.conditional_forecast,'conf_sig') || isempty(options_cond_fcst.conditional_forecast.conf_sig) + options_cond_fcst.conditional_forecast.conf_sig = .8; end if isequal(options_cond_fcst.parameter_set,'calibration') @@ -228,7 +228,7 @@ end mFORCS1 = mean(FORCS1,3); mFORCS1_shocks = mean(FORCS1_shocks,3); -tt = (1-options_cond_fcst.conf_sig)/2; +tt = (1-options_cond_fcst.conditional_forecast.conf_sig)/2; t1 = round(options_cond_fcst.replic*tt); t2 = round(options_cond_fcst.replic*(1-tt)); diff --git a/matlab/simultxdet.m b/matlab/simultxdet.m index ca34c596c703bc6b230212756764f1ecac9f768c..2de1bb4b7ff19281b59e4978098b720799a89c07 100644 --- a/matlab/simultxdet.m +++ b/matlab/simultxdet.m @@ -134,7 +134,7 @@ for i=1:iter sigma_y = sigma_y+sigma_u; end -fact = norminv((1-options_.conf_sig)/2,0,1); +fact = norminv((1-options_.forecasts.conf_sig)/2,0,1); int_width = zeros(iter,endo_nbr); for i=1:nvar