Problem with identification analysis before the estimation?

Dear Professor Pfeifer,

When I was doing the identification analysis before the bayesian estimation, I add the identification command in my code and get the error message as bellow:

There are NaN’s in the theoretical moments: make sure that for non-stationary models stationary transformations of non-stationary observables are used for checking identification. [TIP: use first differences].

I’m confused about this error because I have already used One-sided HP filter to deal with log data, and the data seems stationary. I searched the forum and did not find the solution. Even I eliminate all the shock but the Taylor Rule and use only one observable, the error still exists.

Would you please give some advise on how to fix it ? Thank you very much.
Following files are related. It’s a model without specified trend and the datas are detrended.
hm.mod (11.7 KB) hm_data.m (10.5 KB) hm_steadystate.m (6.8 KB)

The identification-command belongs after the estimated_params-block.