Page 1 of 1

Difference between smoothedvariables and updatedvariables

PostPosted: Thu Aug 20, 2015 10:37 pm
by Seungcheol Lee
Hello.

I like to know the difference between oo_.smoothedvariables and oo_.updatedvariables.

It's difficult to know the exact difference through manual.

Specifically, which value should I use for estimation of endogenous variables (for example, output gap) ?

Thank you for your help in advance!

Re: Difference between smoothedvariables and updatedvariable

PostPosted: Fri Aug 21, 2015 6:07 am
by jpfeifer
What do you not understand? As the manual says, smoothed objects are the
expected value of variables and shocks given
the information available in all observations up to the final date (E_T (y_t)).

They use the full sample information for inference.

In contrast, the updated variables
contains the estimation of the expected value of
variables given the information available at the current date (E_t (y_t)).

They are basically real-time estimates as they only use information up to the current date t.

Most of the time, people are interested in smoothed values. You want to know the output gap in the past, using all available information to estimate it.

Re: Difference between smoothedvariables and updatedvariable

PostPosted: Fri Aug 21, 2015 8:10 am
by Seungcheol Lee
Dear Jpfeifer

Thank you so much!

Best
Seungcheol