heterogeneity_dimension households; var(heterogeneity=households) c // consumption n // labor supply ns // effective labor supply a // assets Va // derivative of value function w.r.t. a ; varexo(heterogeneity=households) e // idiosyncratic efficiency ; var Y L w pi Div Tax r ; varexo G markup rstar; shocks; var G; stderr 0.01; var markup; stderr 0.01; var rstar; stderr 0.01; end; parameters beta vphi eis frisch rho_e sig_e rho_Z sig_Z mu kappa phi Z B r_ss ; model(heterogeneity=households); // Euler equation with borrowing constraint c^(-1/eis) - beta * Va(+1) = 0 ⟂ a >= 0; // Budget constraint (1 + r) * a(-1) + w * n * e + (Div-Tax) * e - c - a; // Envelope condition Va = (1 + r) * c^(-1/eis); // Intratemporal FOC for labor supply vphi*n^(1/frisch) - w*e*c^(-1/eis); // Definition of the effective labor supply ns = n * e; end; model; // Firm L - Y / Z; Div - (Y - w * L - mu / (mu - 1) / (2 * kappa) * log(1 + pi)^2 * Y); // Monetary policy (Taylor rule) (1 + r_ss + rstar(-1) + phi * pi(-1)) / (1 + pi) - 1 - r; // Fiscal policy Tax - (r * B) - G; // NKPC kappa * (w / Z - 1 / mu) + Y(+1)/Y * log(1 + pi(+1)) / (1 + r(+1)) + markup - log(1 + pi); // Market clearing SUM(a) - B; // asset market sum(ns) - L; // labor market end; mat_file = 'hank_1a_ssj.mat'; if ~exist(mat_file, 'file') url = sprintf('https://dynare.org/test-data/%s', mat_file); websave(mat_file, url); end load(mat_file); param_names = fieldnames(steady_state.params); for i=1:numel(param_names) param = param_names{i}; set_param_value(param, steady_state.params.(param)); end % Initialize steady state once for all tests heterogeneity_load_steady_state(filename = hank_1a_ssj); % Solve once for all tests heterogeneity_solve; verbatim; testFailed = 0; testResults = []; % Array to collect all test results % Test the solve routine and SSJ comparisons (validation only) [testFailed, result] = run_test('heterogeneity.solve functionality and SSJ validation', @() test_solve_functionality(M_, oo_, G, true), testFailed); testResults = [testResults; result]; % Test the solve routine with custom impulse responses (calls solve with custom impulses) [testFailed, result] = run_test('heterogeneity.solve with custom impulse responses', @() test_solve_with_custom_impulse_responses(M_, options_, oo_, x_hat_dash, G, J, F, curlyDs, curlyYs, a_i, a_pi), testFailed); testResults = [testResults; result]; % Test heterogeneity.simulate function (reuse initialized and solved model) [testFailed, testResults] = test_simulate_stochastic(M_, options_, oo_, steady_state, testFailed, testResults); % Print test summary print_test_summary(testResults); if testFailed > 0 error('Some unit tests failed!'); end end;