Strange IRF shape

I make a medium size NK model with sticky price and wage and habit formation and investment adjustment cost.but i find the result is not very correct with the standard one
i do not know what is wrong .please help me check the code .thank you ~
medial.mod (1.37 KB)

What exactly is your problem?

thanks for answering me .my problem is that in the tfp shock irf .output can not return to the steady state and some variables also show this feature in IST shock.i want to know it is because of the fault of model or other reasons to produce this feature. thank you

All IRFs return, but it takes some time. Set

stoch_simul(irf=1000);
to see this

thank you so much!it is very helpful to know this .but i still have a small question. is it normal to take such a long time to return to steady state?

It is not completely unusual. It very much depends on the frictions in the model and the parameterization.