Backus et al. 1994

Hi,
I’m trying to write a program for Backus et al. 1994 but I get this error: Impossible to find the steady state. Either the model doesn’t have a unique steady state of the gues
I dont understand where my mistake is. I wrote everything in levels, I calculated the steady state and I used it as initial guess. Could someone have a look at my file? manythanks
backusetal.mod (1.83 KB)

It is log(a1), log(a2), that follow a linear autocorrelated process, not a1 and a2 (in the last 2 equations)

m