function [BETA,THETA,D,ETA,B,A,gl,gh,phig,zl,zh,phiz,ALFA,TM]=param(MU1) %[BETA,THETA,D,ETA,B,A,gl,gh,phig,zl,zh,phiz,ALFA,TM]=param(MU1) computes the parameters of the model %Input: MU (the markup) %(c) Stephanie Schmitt-Grohe and Martin Uribe ALFA = 1; %fracton of s(v) not rebated BETA = 1/1.04; H=.2; %steady state effort SG=.2; %steady state consumption SB = 0.44; %Transaction cost technology s(v) A=0.0111; B=0.07524; %this assumes s(v) =A v + B/v -2sqrt(AB); see RATS program data/money.prg PAI = 1.042; %Avg US GDP deflator since 1960 %Stochastic processes gl = SG*H*(350/(350+402)*2); gh = SG*H*(402/(350+402)*2); phig=.95; %prob(gh|gh)=prob(gl|gl) %These moments are not needed but for info mean_g = (gh+gl)/2; std_g = gh-mean_g; acov_g = (2*phig-1)*std_g^2; acorr_g = 2*phig-1; zl=1-.04; zh=1+.04; phiz=.91; %prob(zh|zh)=prob(zl|zl) %These moments are not needed but for info mean_z = (zh+zl)/2; std_z = zh-mean_z; acov_z = (2*phiz-1)*std_z^2; acorr_z = 2*phiz-1; %IMPLIED PARAMETERS MU = 1.2; ETA=MU/(1-MU); %MARKUP=MU=ETA/(1+ETA); SC = (1-SG); %consumption share in GDP %CALIBRATION OF THETA %We assume that %U(c,h)=log c + THETA log (1-h) R = PAI/BETA; %Gross nominal interest rate V = sqrt(B/A+(R-1)/R/A); S=A*V+B/V-2*sqrt(A*B); SP=A-B/V^2; GAMA=1+S+V*SP; TAU = -MU*(SB*(1-R/PAI) - SC/V/(1+ALFA*S)*(1/PAI-1) - SG); THETA = (1-TAU)/ MU / SC * (1-H)/H *(1+ALFA*S)/GAMA; %SB = (1-R/PAI)^(-1) * (SC/V/PAI/(1+S)+SG-TAU/MU-SC/V/(1+S)) SD = (R*SB+SC/(1+ALFA*S)/V)/PAI; D = H * SD; % Total gov't liabilities ETA=MU1/(1-MU1); %Construct Transition Matrix for Markov processes of g & z % STATES %(gh,zh)=1 %(gh,zl)=2; %(gl,zh)=3; %(gl,zl)=4; TM = [phig*phiz phig*(1-phiz) (1-phig)*phiz (1-phig)*(1-phiz); phig*(1-phiz) phig*phiz (1-phig)*(1-phiz) (1-phig)*phiz; (1-phig)*phiz (1-phig)*(1-phiz) phig*phiz phig*(1-phiz); (1-phig)*(1-phiz) (1-phig)*phiz phig*(1-phiz) phig*phiz];