Negative values after interpolation and seasonal adjustment

For questions regarding programming in the EViews programming language.

Moderators: EViews Gareth, EViews Jason, EViews Moderator, EViews Matt

djackson
Posts: 8
Joined: Fri Jun 19, 2015 1:33 am

Negative values after interpolation and seasonal adjustment

Postby djackson » Wed Mar 31, 2021 9:13 am

Two separate but similar questions.

I have some data that I'm interpolating (using quadratic match average), but some series have negative values after the interpolation. I want to remove the negative values but preserve the annual average of the interpolated data. Currently, I'm changing quarters with negative values to equal the annual value in all quarters later down the line, so I was wondering how to do that in Eviews. Other suggestions are welcome.

I did try linear last match interpolation, which interpolated without creating any negative values, but I'd like to avoid changing the quarterly profile too much if possible.

Similarly, I have some other data that I'm seasonally adjusting (using x12a), which has negative values after adjustment. Again, I'd like to remove the negative values while preserving the annual average and as much of the adjustment as can be preserved.

EViews Gareth
Fe ddaethom, fe welon, fe amcangyfrifon
Posts: 13306
Joined: Tue Sep 16, 2008 5:38 pm

Re: Negative values after interpolation and seasonal adjustment

Postby EViews Gareth » Wed Mar 31, 2021 9:50 am

I think this is first an economic question (i.e. what exactly do you want to do with those negative values), and then a programming question.
Follow us on Twitter @IHSEViews

djackson
Posts: 8
Joined: Fri Jun 19, 2015 1:33 am

Re: Negative values after interpolation and seasonal adjustment

Postby djackson » Thu Apr 01, 2021 4:36 am

I'd like to remove them from the interpolated/adjusted data while keeping the annual average of the data the same. I assume this would involve proportioning the quarters differently in years with negative values.


Return to “Programming”

Who is online

Users browsing this forum: No registered users and 16 guests