Problems with assessing in-sample fit


I am working with a small open economy model for a country in a monetary union. There are no technical problems in solving/estimating the model and IRFs are fine, but I have a problem when I come to compute the in-sample fit of the model, by plotting the one period ahead forecast against the observed data. However, the ‘Filtered data’ series, which constitutes the one-step ahead prediction (derived from _oo) is essentially made up of zeros, with the exception of the last value. What is the meaning of this?

I am attaching both the .mod file and the data file.

data.m (4.62 KB)
soe_MU.mod (6 KB)