function oo_ = mcmc_diagnostics(options_, estim_params_, M_, oo_) % function oo_ = mcmc_diagnostics(options_, estim_params_, M_, oo_) % Computes convergence tests % % INPUTS % options_ [structure] Dynare options structure % estim_params_ [structure] Dynare estimation parameter structure % M_ [structure] Dynare model structure % oo_ [structure] Dynare results structure % % OUTPUTS % oo_ [structure] % % SPECIAL REQUIREMENTS % none % % PARALLEL CONTEXT % See the comment in posterior_sampler.m funtion. % Copyright © 2005-2023 Dynare Team % % This file is part of Dynare. % % Dynare is free software: you can redistribute it and/or modify % it under the terms of the GNU General Public License as published by % the Free Software Foundation, either version 3 of the License, or % (at your option) any later version. % % Dynare is distributed in the hope that it will be useful, % but WITHOUT ANY WARRANTY; without even the implied warranty of % MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the % GNU General Public License for more details. % % You should have received a copy of the GNU General Public License % along with Dynare. If not, see . graphFolder = CheckPath('graphs',M_.dname); latexFolder = CheckPath('latex',M_.dname); MetropolisFolder = CheckPath('metropolis',M_.dname); ModelName = M_.fname; TeX = options_.TeX; record=load_last_mh_history_file(MetropolisFolder, ModelName); NumberOfMcFilesPerBlock = record.LastFileNumber; [nblck, npar] = size(record.LastParameters); npardisp = options_.convergence.brooksgelman.plotrows; % check if all the mh files are available. issue_an_error_message = 0; for b = 1:nblck nfiles = length(dir([MetropolisFolder ,filesep, ModelName '_mh*_blck' num2str(b) '.mat'])); if ~isequal(NumberOfMcFilesPerBlock,nfiles) issue_an_error_message = 1; fprintf('The number of MCMC files in chain %u is %u while the mh-history files indicate that we should have %u MCMC files per chain!\n',b, nfiles, NumberOfMcFilesPerBlock); end end if issue_an_error_message error('mcmc_diagnostics: I cannot proceed because some MCMC files are missing. Check your MCMC files...!'); end % compute inefficiency factor FirstLine = record.KeepedDraws.FirstLine; TotalNumberOfMhFiles = sum(record.MhDraws(:,2)); TotalNumberOfMhDraws = sum(record.MhDraws(:,1)); FirstMhFile = record.KeepedDraws.FirstMhFile; NumberOfDraws = TotalNumberOfMhDraws-floor(options_.mh_drop*TotalNumberOfMhDraws); param_name = {}; param_name_tex = {}; Ifac=NaN(nblck,npar); for jj = 1:npar if options_.TeX [par_name_temp, par_name_tex_temp] = get_the_name(jj, options_.TeX, M_,estim_params_, options_.varobs); param_name = vertcat(param_name, par_name_temp); par_name_tex_temp = strrep(par_name_tex_temp,'$',''); param_name_tex = vertcat(param_name_tex, par_name_tex_temp); else par_name_temp = get_the_name(jj, options_.TeX, M_, estim_params_, options_.varobs); param_name = vertcat(param_name, par_name_temp); end Draws = GetAllPosteriorDraws(options_, M_.dname, M_.fname, jj, FirstMhFile, FirstLine, TotalNumberOfMhFiles, NumberOfDraws, nblck); Draws = reshape(Draws, [NumberOfDraws nblck]); Nc = min(1000, NumberOfDraws/2); for ll = 1:nblck Ifac(ll,jj) = mcmc_ifac(Draws(:,ll), Nc); end tmp = num2cell(Ifac(:,jj)); end my_title='MCMC Inefficiency factors per block'; IFAC_header = {'Parameter'}; IFAC_header_tex = {'Parameter'}; for j = 1:nblck IFAC_header = vertcat(IFAC_header, ['Block ' int2str(j)]); IFAC_header_tex = vertcat(IFAC_header_tex, ['Block~' int2str(j)]); end lh = cellofchararraymaxlength(param_name)+2; dyntable(options_, my_title, IFAC_header, param_name, Ifac', lh, 12, 3); skipline() if options_.TeX dyn_latex_table(M_, options_, my_title, 'MCMC_inefficiency_factors', IFAC_header_tex, param_name_tex, Ifac', lh, 12, 3); end record.InefficiencyFactorsPerBlock = Ifac; update_last_mh_history_file(MetropolisFolder, ModelName, record); PastDraws = sum(record.MhDraws,1); NumberOfDraws = PastDraws(1); if ~strcmp(options_.posterior_sampler_options.posterior_sampling_method,'slice') && NumberOfDraws<=2000 warning('MCMC convergence diagnostics are not computed because the total number of iterations is not bigger than 2000!'); return end convergence_diagnostics_geweke=zeros(npar,4+2*length(options_.convergence.geweke.taper_steps)); if any(options_.convergence.geweke.geweke_interval<0) || any(options_.convergence.geweke.geweke_interval>1) || length(options_.convergence.geweke.geweke_interval)~=2 ... || (options_.convergence.geweke.geweke_interval(2)-options_.convergence.geweke.geweke_interval(1)<0) fprintf('\nCONVERGENCE DIAGNOSTICS: Invalid option for geweke_interval. Using the default of [0.2 0.5].\n') options_.convergence.geweke.geweke_interval=[0.2 0.5]; end first_obs_begin_sample = max(1,ceil(options_.mh_drop*NumberOfDraws)); last_obs_begin_sample = first_obs_begin_sample+round(options_.convergence.geweke.geweke_interval(1)*NumberOfDraws*(1-options_.mh_drop)); first_obs_end_sample = first_obs_begin_sample+round(options_.convergence.geweke.geweke_interval(2)*NumberOfDraws*(1-options_.mh_drop)); param_name = {}; if options_.TeX param_name_tex = {}; end for jj=1:npar if options_.TeX [param_name_temp, param_name_tex_temp] = get_the_name(jj, options_.TeX, M_, estim_params_, options_.varobs); param_name_tex = vertcat(param_name_tex, strrep(param_name_tex_temp, '$','')); param_name = vertcat(param_name, param_name_temp); else param_name_temp = get_the_name(jj, options_.TeX, M_,estim_params_, options_.varobs); param_name = vertcat(param_name, param_name_temp); end end datamat=NaN(npar,3+length(options_.convergence.geweke.taper_steps),nblck); %remove stale results as it will cause assignment problems if options_.convergence.rafterylewis.indicator && isfield(oo_,'convergence') && isfield(oo_.convergence,'raftery_lewis') oo_.convergence=rmfield(oo_.convergence,'raftery_lewis'); end for block_iter=1:nblck fprintf('\n\nConvergence diagnostics results for chain %u.\n',block_iter); fprintf('\nGeweke (1992) Convergence Tests, based on means of draws %d to %d vs %d to %d for chain %u.\n',first_obs_begin_sample,last_obs_begin_sample,first_obs_end_sample,NumberOfDraws,block_iter); fprintf('p-values are for Chi2-test for equality of means.\n'); Geweke_header = {'Parameter'; 'Post. Mean'; 'Post. Std'; 'p-val No Taper'}; for ii = 1:length(options_.convergence.geweke.taper_steps) Geweke_header = vertcat(Geweke_header, ['p-val ' num2str(options_.convergence.geweke.taper_steps(ii)),'% Taper']); end for jj=1:npar startline=0; for n = 1:NumberOfMcFilesPerBlock load([MetropolisFolder '/' ModelName '_mh',int2str(n),'_blck',num2str(block_iter),'.mat'],'x2'); nx2 = size(x2,1); param_draws(startline+(1:nx2),1) = x2(:,jj); startline = startline + nx2; end [results_vec, results_struct] = geweke_moments(param_draws,options_); convergence_diagnostics_geweke(jj,:)=results_vec; param_draws1 = param_draws(first_obs_begin_sample:last_obs_begin_sample,:); param_draws2 = param_draws(first_obs_end_sample:end,:); [results_vec1] = geweke_moments(param_draws1,options_); [results_vec2] = geweke_moments(param_draws2,options_); results_struct = geweke_chi2_test(results_vec1,results_vec2,results_struct,options_); oo_.convergence.geweke(block_iter).(param_name{jj}) = results_struct; datamat(jj,:,block_iter)=[results_struct.posteriormean,results_struct.posteriorstd,results_struct.prob_chi2_test]; end lh = size(param_name,2)+2; dyntable(options_, '', Geweke_header, param_name, datamat(:,:,block_iter), lh, 12, 3); if options_.TeX Geweke_tex_header = {'Parameter'; 'Mean'; 'Std'; 'No\ Taper'}; additional_header = {[' & \multicolumn{2}{c}{Posterior} & \multicolumn{',num2str(1+length(options_.convergence.geweke.taper_steps)),'}{c}{p-values} \\'], ['\cmidrule(r{.75em}){2-3} \cmidrule(r{.75em}){4-',num2str(4+length(options_.convergence.geweke.taper_steps)),'}']}; for ii=1:length(options_.convergence.geweke.taper_steps) Geweke_tex_header = vertcat(Geweke_tex_header, [num2str(options_.convergence.geweke.taper_steps(ii)),'\%%\ Taper']); end headers = Geweke_tex_header; lh = cellofchararraymaxlength(param_name_tex)+2; my_title=sprintf('Geweke (1992) Convergence Tests, based on means of draws %d to %d vs %d to %d for chain %u. p-values are for $\\\\chi^2$-test for equality of means.',first_obs_begin_sample,last_obs_begin_sample,first_obs_end_sample,NumberOfDraws,block_iter); dyn_latex_table(M_, options_, my_title, ['geweke_block_' num2str(block_iter)], headers, param_name_tex, datamat(:,:,block_iter), lh, 12, 4, additional_header); end skipline(2); if options_.convergence.rafterylewis.indicator if any(options_.convergence.rafterylewis.qrs<0) || any(options_.convergence.rafterylewis.qrs>1) || length(options_.convergence.rafterylewis.qrs)~=3 ... || (options_.convergence.rafterylewis.qrs(1)-options_.convergence.rafterylewis.qrs(2)<=0) fprintf('\nInvalid option for raftery_lewis_qrs. Using the default of [0.025 0.005 0.95].\n'); options_.convergence.rafterylewis.qrs=[0.025 0.005 0.95]; end Raftery_Lewis_q=options_.convergence.rafterylewis.qrs(1); Raftery_Lewis_r=options_.convergence.rafterylewis.qrs(2); Raftery_Lewis_s=options_.convergence.rafterylewis.qrs(3); oo_.convergence.raftery_lewis(block_iter) = raftery_lewis(x2,Raftery_Lewis_q,Raftery_Lewis_r,Raftery_Lewis_s); my_title=sprintf('Raftery/Lewis (1992) Convergence Diagnostics, based on quantile q=%4.3f with precision r=%4.3f with probability s=%4.3f for chain %u.',Raftery_Lewis_q,Raftery_Lewis_r,Raftery_Lewis_s,block_iter); headers = {'Variables'; 'M (burn-in)'; 'N (req. draws)'; 'N+M (total draws)'; 'k (thinning)'}; raftery_data_mat=[oo_.convergence.raftery_lewis(block_iter).M_burn,oo_.convergence.raftery_lewis(block_iter).N_prec,oo_.convergence.raftery_lewis(block_iter).N_total,oo_.convergence.raftery_lewis(block_iter).k_thin]; raftery_data_mat=[raftery_data_mat; max(raftery_data_mat,[],1)]; labels_Raftery_Lewis = vertcat(param_name, 'Maximum'); lh = cellofchararraymaxlength(labels_Raftery_Lewis)+2; dyntable(options_, my_title, headers, labels_Raftery_Lewis, raftery_data_mat, lh, 10, 0); if options_.TeX labels_Raftery_Lewis_tex = vertcat(param_name_tex, 'Maximum'); lh = cellofchararraymaxlength(labels_Raftery_Lewis_tex)+2; dyn_latex_table(M_, options_, my_title, ['raftery_lewis_' num2str(block_iter)], headers, labels_Raftery_Lewis_tex, raftery_data_mat, lh, 10, 0); end end end for block_iter=1:nblck oo_.convergence.raftery_lewis(block_iter).parameter_names=param_name; end if nblck==1 return end if strcmp(options_.posterior_sampler_options.posterior_sampling_method,'slice') && NumberOfDraws<2000 Origin = 1; StepSize = 1; else Origin = 1000; StepSize = ceil((NumberOfDraws-Origin)/100);% So that the computational time does not end ALPHA = 0.2; % increase too much with the number of simulations. time = 1:NumberOfDraws; xx = Origin:StepSize:NumberOfDraws; NumberOfLines = length(xx); if NumberOfDraws < Origin fprintf('The number of simulations is too small to compute the MCMC convergence diagnostics.\n'); return end if TeX && any(strcmp('eps',cellstr(options_.graph_format))) fidTeX = fopen([latexFolder '/' ModelName '_UnivariateDiagnostics.tex'],'w'); fprintf(fidTeX,'%% TeX eps-loader file generated by mcmc_diagnostics.m (Dynare).\n'); fprintf(fidTeX,['%% ' datestr(now,0) '\n']); fprintf(fidTeX,' \n'); end fprintf('Univariate convergence diagnostic, Brooks and Gelman (1998):\n'); % The mandatory variables for local/remote parallel % computing are stored in localVars struct. localVars.MetropolisFolder = MetropolisFolder; localVars.nblck = nblck; localVars.NumberOfMcFilesPerBlock = NumberOfMcFilesPerBlock; localVars.Origin = Origin; localVars.StepSize = StepSize; localVars.mh_drop = options_.mh_drop; localVars.NumberOfDraws = NumberOfDraws; localVars.NumberOfLines = NumberOfLines; localVars.time = time; localVars.M_ = M_; % Like sequential execution! if isnumeric(options_.parallel) fout = mcmc_diagnostics_core(localVars,1,npar,0); UDIAG = fout.UDIAG; clear fout % Parallel execution! else if ~isempty(M_.bvar) ModelName = [ModelName '_bvar']; end NamFileInput={[M_.dname '/metropolis/'],[ModelName '_mh*_blck*.mat']}; [fout, ~, totCPU] = masterParallel(options_.parallel, 1, npar,NamFileInput,'mcmc_diagnostics_core', localVars, [], options_.parallel_info); UDIAG = fout(1).UDIAG; for j=2:totCPU UDIAG = cat(3,UDIAG ,fout(j).UDIAG); end end UDIAG(:,[2 4 6],:) = UDIAG(:,[2 4 6],:)/nblck; skipline() clear pmet temp moyenne CSUP CINF csup cinf n linea iter tmp; pages = floor(npar/npardisp); % changed from 3 to npardisp k = 0; for i = 1:pages hh_fig = dyn_figure(options_.nodisplay,'Name','MCMC univariate convergence diagnostic (Brooks and Gelman,1998)'); boxplot = 1; for j = 1:npardisp % Loop over parameters %npardisp instead of 3 k = k+1; [nam,namtex] = get_the_name(k,TeX,M_,estim_params_,options_.varobs); for crit = 1:3% Loop over criteria if crit == 1 plt1 = UDIAG(:,1,k); plt2 = UDIAG(:,2,k); namnam = [nam , ' (Interval)']; elseif crit == 2 plt1 = UDIAG(:,3,k); plt2 = UDIAG(:,4,k); namnam = [nam , ' (m2)']; elseif crit == 3 plt1 = UDIAG(:,5,k); plt2 = UDIAG(:,6,k); namnam = [nam , ' (m3)']; end subplot(npardisp,3,boxplot); %Added more rows to display more variables plot(xx,plt1,'-b'); % Pooled hold on; plot(xx,plt2,'-r'); % Within (mean) hold off; xlim([xx(1) xx(NumberOfLines)]) if TeX title(namtex,'interpreter','latex') else title(namnam,'Interpreter','none') end boxplot = boxplot + 1; end end dyn_saveas(hh_fig,[graphFolder '/' ModelName '_udiag' int2str(i)],options_.nodisplay,options_.graph_format); if TeX && any(strcmp('eps',cellstr(options_.graph_format))) fprintf(fidTeX,'\\begin{figure}[H]\n'); fprintf(fidTeX,'\\centering \n'); fprintf(fidTeX,'\\includegraphics[width=%2.2f\\textwidth]{%s_udiag%s}\n',options_.figures.textwidth*min((boxplot-1)/3,1),[graphFolder '/' ModelName],int2str(i)); fprintf(fidTeX,'\\caption{Univariate convergence diagnostics for the Metropolis-Hastings.\n'); fprintf(fidTeX,'The first, second and third columns are respectively the criteria based on\n'); fprintf(fidTeX,'the eighty percent interval, the second and third moments.}'); fprintf(fidTeX,'\\label{Fig:UnivariateDiagnostics:%s}\n',int2str(i)); fprintf(fidTeX,'\\end{figure}\n'); fprintf(fidTeX,'\n'); end end reste = npar-k; if reste if reste == 1 nr = 3; nc = 1; else % Conditional for additional rows (variables) when not towards the end of the loop nr = npardisp; nc = 3; end hh_fig = dyn_figure(options_.nodisplay,'Name','MCMC univariate convergence diagnostic (Brooks and Gelman, 1998)'); boxplot = 1; for j = 1:reste k = k+1; [nam,namtex] = get_the_name(k,TeX,M_,estim_params_,options_.varobs); for crit = 1:3 if crit == 1 plt1 = UDIAG(:,1,k); plt2 = UDIAG(:,2,k); namnam = [nam , ' (Interval)']; if TeX namnamtex = [namtex , ' (Interval)']; end elseif crit == 2 plt1 = UDIAG(:,3,k); plt2 = UDIAG(:,4,k); namnam = [nam , ' (m2)']; if TeX namnamtex = [namtex , ' (m2)']; end elseif crit == 3 plt1 = UDIAG(:,5,k); plt2 = UDIAG(:,6,k); namnam = [nam , ' (m3)']; if TeX namnamtex = [namtex , ' (m3)']; end end subplot(nr,nc,boxplot); plot(xx,plt1,'-b'); % Pooled hold on; plot(xx,plt2,'-r'); % Within (mean) hold off; xlim([xx(1) xx(NumberOfLines)]); if TeX title(namnamtex,'Interpreter','latex'); else title(namnam,'Interpreter','none'); end boxplot = boxplot + 1; end end dyn_saveas(hh_fig,[ graphFolder '/' ModelName '_udiag' int2str(pages+1)],options_.nodisplay,options_.graph_format); if TeX && any(strcmp('eps',cellstr(options_.graph_format))) fprintf(fidTeX,'\\begin{figure}[H]\n'); fprintf(fidTeX,'\\centering \n'); fprintf(fidTeX,'\\includegraphics[width=%2.2f\\textwidth]{%s_udiag%s}\n',options_.figures.textwidth*min((boxplot-1)/nc,1),[graphFolder '/' ModelName],int2str(pages+1)); if reste == 2 fprintf(fidTeX,'\\caption{Univariate convergence diagnostics for the Metropolis-Hastings.\n'); fprintf(fidTeX,'The first, second and third columns are respectively the criteria based on\n'); fprintf(fidTeX,'the eighty percent interval, the second and third moments.}'); elseif reste == 1 fprintf(fidTeX,'\\caption{Univariate convergence diagnostics for the Metropolis-Hastings.\n'); fprintf(fidTeX,'The first, second and third rows are respectively the criteria based on\n'); fprintf(fidTeX,'the eighty percent interval, the second and third moments.}'); end fprintf(fidTeX,'\\label{Fig:UnivariateDiagnostics:%s}\n',int2str(pages+1)); fprintf(fidTeX,'\\end{figure}\n'); fprintf(fidTeX,'\n'); fprintf(fidTeX,'%% End Of TeX file.'); fclose(fidTeX); end end % if reste > 0 clear UDIAG; % % Multivariate diagnostic. % if TeX && any(strcmp('eps',cellstr(options_.graph_format))) fidTeX = fopen([latexFolder '/' ModelName '_MultivariateDiagnostics.tex'],'w'); fprintf(fidTeX,'%% TeX eps-loader file generated by mcmc_diagnostics.m (Dynare).\n'); fprintf(fidTeX,['%% ' datestr(now,0) '\n']); fprintf(fidTeX,' \n'); end tmp = zeros(NumberOfDraws*nblck,3); MDIAG = zeros(NumberOfLines,6); for b = 1:nblck startline = 0; for n = 1:NumberOfMcFilesPerBlock load([MetropolisFolder '/' ModelName '_mh',int2str(n),'_blck' int2str(b) '.mat'],'logpo2'); nlogpo2 = size(logpo2,1); tmp((b-1)*NumberOfDraws+startline+(1:nlogpo2),1) = logpo2; startline = startline+nlogpo2; end end clear logpo2; tmp(:,2) = kron(transpose(1:nblck),ones(NumberOfDraws,1)); tmp(:,3) = kron(ones(nblck,1),time'); tmp = sortrows(tmp,1); ligne = 0; for iter = Origin:StepSize:NumberOfDraws ligne = ligne+1; linea = ceil(options_.mh_drop*iter); n = iter-linea+1; cinf = max(1,round(n*ALPHA/2)); csup = round(n*(1-ALPHA/2)); CINF = max(1,round(nblck*n*ALPHA/2)); CSUP = round(nblck*n*(1-ALPHA/2)); temp = tmp(find((tmp(:,3)>=linea) & (tmp(:,3)<=iter)),1:2); MDIAG(ligne,1) = temp(CSUP,1)-temp(CINF,1); moyenne = mean(temp(:,1));%% Pooled mean. MDIAG(ligne,3) = sum((temp(:,1)-moyenne).^2)/(nblck*n-1); MDIAG(ligne,5) = sum(abs(temp(:,1)-moyenne).^3)/(nblck*n-1); for i=1:nblck pmet = temp(find(temp(:,2)==i)); MDIAG(ligne,2) = MDIAG(ligne,2) + pmet(csup,1)-pmet(cinf,1); moyenne = mean(pmet,1); %% Within mean. MDIAG(ligne,4) = MDIAG(ligne,4) + sum((pmet(:,1)-moyenne).^2)/(n-1); MDIAG(ligne,6) = MDIAG(ligne,6) + sum(abs(pmet(:,1)-moyenne).^3)/(n-1); end end MDIAG(:,[2 4 6],:) = MDIAG(:,[2 4 6],:)/nblck; hh_fig = dyn_figure(options_.nodisplay,'Name','Multivariate convergence diagnostic'); boxplot = 1; for crit = 1:3 if crit == 1 plt1 = MDIAG(:,1); plt2 = MDIAG(:,2); namnam = 'Interval'; elseif crit == 2 plt1 = MDIAG(:,3); plt2 = MDIAG(:,4); namnam = 'm2'; elseif crit == 3 plt1 = MDIAG(:,5); plt2 = MDIAG(:,6); namnam = 'm3'; end subplot(3,1,boxplot); plot(xx,plt1,'-b'); % Pooled hold on plot(xx,plt2,'-r'); % Within (mean) hold off xlim([xx(1) xx(NumberOfLines)]) title(namnam,'Interpreter','none'); boxplot = boxplot + 1; end dyn_saveas(hh_fig,[ graphFolder '/' ModelName '_mdiag'],options_.nodisplay,options_.graph_format); if TeX && any(strcmp('eps',cellstr(options_.graph_format))) fprintf(fidTeX,'\\begin{figure}[H]\n'); fprintf(fidTeX,'\\centering \n'); fprintf(fidTeX,'\\includegraphics[width=0.8\\textwidth]{%s_mdiag}\n',[graphFolder '/' ModelName]); fprintf(fidTeX,'\\caption{Multivariate convergence diagnostics for the Metropolis-Hastings.\n'); fprintf(fidTeX,'The first, second and third rows are respectively the criteria based on\n'); fprintf(fidTeX,'the eighty percent interval, the second and third moments. The different \n'); fprintf(fidTeX,'parameters are aggregated using the posterior kernel.}'); fprintf(fidTeX,'\\label{Fig:MultivariateDiagnostics}\n'); fprintf(fidTeX,'\\end{figure}\n'); fprintf(fidTeX,'\n'); fprintf(fidTeX,'%% End Of TeX file.'); fclose(fidTeX); end