Dear Johannes,
Thank you for your helpful guidance last time, I am grateful.
I have a question about historical decomposition, does historical decomposition of an observable variable represent nowcast errors of observable variable?
Thank you very much and look forward to hearing from you.
Best regards,
Jesse
Hi Jesse,
the historical shock decomposition is based on the Kalman Smoother.
This implies that you use information up to T. Hence, I would guess that your errors are not true nowcast errors, since the later are a concept of forecasting not yet observed variables at time t given information at t. But, maybe the output from the Kalman Filter could be thought of as nowcasts/nowcast errors.