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,
Error in dynare_estimation_1 (line 474)
Error in dynare_estimation (line 105)
Error in neoclassical.driver (line 240)
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.