Smoothed Shocks

[stl84]

I have a question about the estimated smoothed shocks.
As far as I know, the smoothed shocks are the best guess based on all observations.

The problem is that I've gotten a large scale (around 1e+04) of smoothed shocks over the initial 20 quarters.
Although it converges to zero after the initial 20 quarters, I was wondering what causes the initial spikes of the smoothed shocks.

I always appreciate your help.
Thanks

[jpfeifer]

Most probably a mistake in your observation equations where you need big shocks to bring a mean 0 variable to the level of the data.

[stl84]

Many thanks.
I got some intuitions from your comments.

Are there any possibilities of absurd initial values of state variables (x_{0|0} and P_{0|0}) or wrong calibrated parameters?

[jpfeifer]

The initial values for the Kalman filter are the steady state and the unconditional variance. If your observation equation is wrong, that starting value will of course be off.