I wonder what jumps in perfect foresight solution mean. I plot the simulated results in forms of the deviations from the initval (the initial steady state of my model). A few variables display jumps in period 2 specifically.
I have the same problem, I know there can be some jumps but they are very severe and almost completely converged back at the following period, so very similar to what siri said. Did you find a solution? I really appreciate any help!