Question about Binding Constraint

I have a set of log linearized equations.

Within the set of log linearized equations, I am trying to place

i_s_hat = max(0, i_r_hat);

where i_s_hat is a % deviation from steady state savings interest rate

and i_r_hat is a % deviation from steady state reserve interest rate

However, when I use stoch_simul command to output impulse response functions, Dynare ignores i_s_hat in the model block.

If Dynare receives negative value of i_r_hat, I want Dynare to give value of 0 to i_s_hat.

So, I have looked for the answers online. What I found is Occbin Toolkit. Is this a toolkit that is appropriate for this situation?



It depends on what you are trying to do. Perturbation solutions are not capable of dealing with occasionally binding constraints. Thus, you need to use toolkits like Occbin or DynareOBC.

1 Like

Thank you for the response Professor Pfeifer,

Oh I just want an impulse response functions to a preference shock.

Hence, instead of stoch_simul(), is simul() also sufficient in this situation?



Dear Professor Pfeifer

I want an impulse response function to a preference shock for a “bounded” case.

To describe more in detail about the “bounded” case,

The steady state of the “bounded” case is that saving interest rate is bounded which means that saving interest rate cannot go down any further. There is a preference shock that decreases the reserve interest rate. There is also a relationship between reserve interest rate and savings interest rate which is the i_s_hat = max(0, i_r_hat); Hence, I want savings interest rate to take a value of 0 whenever i_r_hat is negative.

I am thinking of running simul() in this situation.



Dear Professor Pfeifer,

Sorry for the massive replies. I have figured out that OccBin is what I want. I will ask questions in future if needed. Thank you for your help!