Stochastic steady state in model equations

Hi everyone,

my question is: is it possible to have the stochastic steady state in the model block?

Say I am writing a Taylor rule such as i_t=i^{\star} + \rho (\pi_t-\pi^{\star}). I would like to have the stochastic steady state of i in i^{\star}. For the deterministic steady state I could write something like int=steady_state(int)+RHO_PI*(infl-steady_state(infl)). But what if I want the stochastic one?

Thanks a lot in advance

No, that is unfortunately not feasible.

1 Like

Thanks a lot! is there a state-of-the-art way to do it? I was thinking of creating an endogenous variable such that averages (for example) the last 20 observations for the interest, like
i_ss=(int(-1)+int(-2)+…+int(-20))/20
I get that 20 periods might not be enough but for now I could not come up with a less convoluted way.

No, you need a parameter that stores the stochastic steady state (SSS). That creates a fixed point problem. The model equations depend on the stochastic steady state, which in turn depends on the model equations containing the parameter. The only way I can think of is setting an initial numerical value, computing the SSS conditional on this value and then updating the parameter value.