mixing anticipated and unanticipated shocks in a simulation

[prom1122]

I would like to create a simulation story, that is to see how 2 different shocks drive the economy in a simulation.
Is it possible to mix anticipated with unanticipated shocks? For example, an anticipated shock hits for 5 consecutive periods, and additionally in period 6 an unanticipated shock comes in.

i have tried to construct the 2 scenarios manually outside Dynare by concatenating the responses from the deterministic simulation, with the IRFs from a stochastic simulation. My problem is that I want the stochastic simulation to be reported as % deviation from the SS of the model, but the simulation in period 6 must use as a baseline the values from period 5 of the deterministic simulation, and NOT the steady state of the model. What I get after concatenating is that the variables return to the baseline of period 5, and not the SS.

Any ideas if this can be implemented in Dynare? or if such a vector can be construct outside in a clever way?

many thanks

[jpfeifer]

Define news shocks and surprise shocks, then set the shock matrix as desired and use the simult_-function for the simulation. An example is at https://github.com/JohannesPfeifer/DSGE_mod/blob/master/RBC_news_shock_model/RBC_news_shock_model.mod