I am writing for signalling a possible bug in the deterministic simulation of the unstable version of Dynare.
In particular: I am simulating a model which starts from an initial steady state and, following a permanent shock, it ends up in a different steady state.
In my mod. file I specified both initial and terminal conditions and Dynare recognizes terminal conditions as the steady state of the model when the command “steady” follows “endval”.
Nevertheless, it seems that Dynare completely ignores the initial conditions: in the simulation, the model is starting and ending in the same steady state.
E.g for any endogenous variable, say ‘c’, c(1)-c(end)=0.
Is there something I am missing? (the mod file is the same posted here: Min () in deterministic simulation)