Dynare forums

Hi,
We are starting to introduce in this computacional world
We are replaying a model of traded and nontraded goods of Wickens (2008) on Octave and the program run the codes, but at the end appear: "error: Impossible to find the steady state. Either the model doesn't have a steady state, there are an infinity of steady state, or the guess values are too far from the solution".
So we try to give another initial values according with the theory, but it don't works.

We do the system of ecuations simple, but the ecuation 14 and 15 require that the shocks EZt and EZn have a special distribution is a normal distribution with media 0, and variance is a matrix of:
(0 delta
delta 0)
but I don't know how introduce a distribution in the program and maybe this cause the before mistake.


So if your can guide some little and tell us what we can do, we apreciate that .


Thanks!

PD: sorry by my english.

1. It looks as if you are trying to determine something endogenously that cannot be determined within the model. Are you sure that all individual prices are uniquely identified? In most models, it is only the relative prices and inflation rates that can be computed from the model. In that case, you need to introduce and arbitrary normalization into the model, like P=1.
2. Use more sensible starting values. You use It>Kt, but you know that It=dt*Kt. Also, you know that Zt=Zn=1.

3. Your last question regarding the distribution: you need a shocks block. See the manual. But this is not the reason of the problem here.