// Simple test of shock_paths with perfect_foresight_with_expectation_errors_{setup,solver} var c k; varexo x; parameters alph gam delt bet aa; alph=0.5; gam=0.5; delt=0.02; bet=0.05; aa=0.5; model; c + k - aa*x*k(-1)^alph - (1-delt)*k(-1); c^(-gam) - (1+bet)^(-1)*(aa*alph*x(+1)*k^(alph-1) + 1 - delt)*c(+1)^(-gam); end; initval; x = 1; k = ((delt+bet)/(1.0*aa*alph))^(1/(alph-1)); c = aa*k^alph-delt*k; end; steady; check; shock_paths; var x; periods 2018Y, 2019Y, 2020Y:end; values 1.2, self.x(-1) + 0.01, initval.x + 0.02; end; shock_paths(learnt_in = 2); var x; periods 2019Y; values prev.x + 0.03; end; shock_paths(learnt_in = 2024Y); var x; periods 2026Y, 2027Y:end; values learnt_in(2019Y).x(+1) + 0.04, learnt_in(2018Y).x(-8); end; perfect_foresight_with_expectation_errors_setup(periods=200, first_simulation_period = 2018Y); perfect_foresight_with_expectation_errors_solver; assert(oo_.exo_simul(3) == 1.2 + 0.01 + 0.03) assert(all(oo_.exo_simul(4:9) == 1 + 0.02)) assert(oo_.exo_simul(10) == 1 + 0.02 + 0.04) assert(oo_.exo_simul(11) == 1.2 + 0.01) assert(all(oo_.exo_simul(12:202) == 1 + 0.02))