why the two standard error not same from practicing dynare?

[lyjessica]

Hi everybody,

I have a problem with the Practicing Dynare in page 46-47.

in page 46, it says:
shocks;
var e_d;
stderr 1;
var e_s;
stderr 1;
end;

but in page 47, when estimate the the parameters using Bayesian techniques, it says:
estimated_params;
a1, gamma_pdf,.5,.5;
d1, gamma_pdf,.2, .5;
stderr e_d, inv_gamma_pdf, 5, 30;
stderr e_s, inv_gamma_pdf, .2, 30;

My problem is why e_d and e_s's stderr are set to 1 in 'shocks', and the two standard errors are set to 30 in 'estimated_params'.

can anybody help me?

thankyou!

Jessica

how_to_use_dynare_to_calibrate_nyu.pdf

(501.32 KiB) Downloaded 137 times

[jpfeifer]

As the model is linear, the shocks are set 1 for simplicity in the shocks command. In the estimation command, the 30 is not the standard deviation of the parameter, but the standard deviation of the prior distribution for the shock standard deviation. These are two different objects (see the manual for the syntax in the estimated_params-block).

[lyjessica]

thanks, I understand it. thanks again.