Hello,

I wanted to replicate the neoclassical model from Bayesian Macroeconometrics by Del Negro and Schorfheide (2010). I wanted to use the similar data to Smets and Wouters (2007) I downloaded the data from the original sources and changed it in excel to get the variables that I want to estimate. It spans from 1960 to 2021. I took the Dynare code from the Schorfheide’s website and only changed the data. I wanted to estimate per capita real output growth and per capita hours index. However, I got the following error:

Error using chol

Matrix must be positive definite with real diagonal.

Error in posterior_sampler_initialization (line 84)

d = chol(vv);

Error in posterior_sampler (line 60)

posterior_sampler_initialization(TargetFun, xparam1, vv,

mh_bounds,dataset_,dataset_info,options_,M_,estim_params_,bayestopt_,oo_);

Error in dynare_estimation_1 (line 474)

posterior_sampler(objective_function,posterior_sampler_options.proposal_distribution,xparam1,posterior_sampler_options,bounds,dataset_,dataset_info,options_,M_,estim_params_,bayestopt_,oo_);

Error in dynare_estimation (line 105)

dynare_estimation_1(var_list,dname);

Error in neoclassical.driver (line 240)

oo_recursive_=dynare_estimation(var_list_);

Error in dynare (line 293)

evalin(‘base’,[fname ‘.driver’]) ;

I think that this happens because the data that I use is different and thus the observation equations are not specified properly. I think the original data was not in per capita terms. Can you please help me? I need any solution as soon as possible. Thank you for the understanding.

neoclassical.mod (2.9 KB)

us1.xlsx (16.8 KB)

Here are the model and excel files that I used.