Page 1 of 1

Problem with historical and smoothed variables

PostPosted: Wed Aug 19, 2015 5:38 am
by econ86
De

Re: Problem with historical and smoothed variables

PostPosted: Wed Aug 19, 2015 8:24 am
by jpfeifer
How are they different? Just a shift in mean? Or different dynamics? The smoothed variables are typically saved as deviations from the mean. If the data mean is not 0, there will appear a shift.

Re: Problem with historical and smoothed variables

PostPosted: Wed Aug 19, 2015 11:50 am
by econ86
De

Re: Problem with historical and smoothed variables

PostPosted: Wed Aug 19, 2015 1:55 pm
by jpfeifer
That looks weird. Please provide the codes to replicate the issue.

Re: Problem with historical and smoothed variables

PostPosted: Fri Aug 21, 2015 9:27 am
by econ86
I

Re: Problem with historical and smoothed variables

PostPosted: Fri Aug 21, 2015 10:51 am
by jpfeifer
There must be something in the model that tightly links these. But I still wonder what happens that there are differences in observed and smoothed series, because this requires measurement error.