capital adjustment costs

[John123]

Dear all,

i want to compare two models, one with quadratic capital adjustment costs and one without. For the latter i set the parameter for the costs (tau_K) zero.
But doings this yields the following error message, that the rank condition isn't verified.

There are 7 eigenvalue(s) larger than 1 in modulus
for 8 forward-looking variable(s)

The rank conditions ISN'T verified!

??? Error using ==> print_info
Blanchard Kahn conditions are not satisfied: indeterminacy

Error in ==> stoch_simul at 46
print_info(info, options_.noprint);

Error in ==> NKM at 210
info = stoch_simul(var_list_);

Error in ==> dynare at 132
evalin('base',fname) ;

i checked the model several times, but can't find any error. It would be great, if you could help me. Thank you

[jpfeifer]

For some parametrizations of real and nominal frictions, capital adjustment costs can be necessary to achieve determinacy: see e.g.
http://www.econ.jhu.edu/pdf/papers/WP490_lubik.pdf
Hence, your model may be completely correct.

[John123]

Thank you very much! I found that the rank condition is fulfilled, if i write in case of tau_K=0 for k and k(+1) K(-1) and k. But that doesn't work, if tau_K is unequal zero and so i'm working there with k and k(+1). Can i compare this two models now?

[jpfeifer]

Sorry, but you cannot simply change the timing of the predetermined capital stock. Apparently you are unsure about the correct timing. I guess the timing is the reason the BK conditions are not satisfied. The line

Code: Select all

STEADY_STATE(y)*y=STEADY_STATE(c)*c +STEADY_STATE(k)*k(+1)-(1-delta)*STEADY_STATE(k)*k;

looks as if you ignore the Dynare "stock at the end of a period" timing convention for predetermined states. Please consult page 10 of the manual. If you fix this problem, you will most probably be analyze the question you have at hand.

[John123]

Thank you very much for your help. I think i understood the timing convention now, unfortunately there is still no solution for my comparision of the two models.