Panel data: Code for copying entry from one year to another year that has no data given that they have the same Name or ID

[Patrick James Lim]

Hi! This is my first post so I apologize in advance if I violate anything.


I'm struggling to put what I want to ask into words so here's an illustration of my current (panel) data:

Name	Year	Description
A	2016	Oil
B	2016	Oil
C	2016	Gas
D	2016	Wood
A	2015
B	2015
C	2015
A	2014
B	2014

So what I want to do is copy Description data to blank observations (in other years) that has exactly the same Name.
Here's what I'm trying to achieve:

Name	Year	Description
A	2016	Oil
B	2016	Oil
C	2016	Gas
D	2016	Wood
A	2015	Oil
B	2015	Oil
C	2015	Gas
A	2014	Oil
B	2014	Oil

The underlined entries under Description are the changes I want to be made.
So what I'm trying to achieve is that a Description entry from a certain Year be copied to another year that has no Description, if the Name is the same.

Notes*
The entries under variable Name are not consistent across years (Name A may appear in Year 2016, 2015, and 2014, but not in 2013 and so on)
I can manually tag them but the data is too big. I'm trying to fill 5000 entries, so I hope there is a way to automate it.

Please let me know if more information is needed. Help would really be appreciated. Thank you!

[Mike Lacy]

I think this will do it for you:

bysort name (description): replace description = description[_N] if (trim(description) == "")

[Nick Cox]

Also, search for mentions of stripolate (SSC) e.g. http://www.statalist.org/forums/foru...-interpolation

[Patrick James Lim]

bysort name (description): replace description = description[_N] if (trim(description) == "")

This made the job like magic.

Also, thanks for sharing the stripolate. Might be really helpful on other works, but I'll need to study it more to use it effectively.

Thank you both!

[Nick Cox]

At a guess, you're really Patrick not Patirck. If so, use the CONTACT US button to arrange a correction with the administrators.

[Patrick James Lim]

Oh I didn't notice! Sorry I guess I was too focused on the problem. Will do, thanks again.