Importing Undated Panel Data

[kruupy]

Hi All,

I have some panel data consisting of 4500 students(STU_NO) from 200(SCH_NO) schools. How do I import this as two dimensional panel data in eviews? - I have tried to import it in many times, however I can only successfully import 1 dimensional using STU_NO as the index. For example I Would like the data to look like:

SCH_NO STU_NO VARIABLES.........
1 1
1 2
1 3
1 4
1 5
2 6
2 7
2 8
2 9
2 10
etc... for each of the 200 schools.

So I would like to run panel regression: Y_ij c X1_ij X2_ij where i = student number and j=school.

Cheers and Thankyou everyone for your help.
Cheers,
Andrew/

[EViews Gareth]

Should be able to open the file in EViews, then use Proc->Restructure to structure it as an undated panel.

[kruupy]

Hi,

I now have the data imported so that STU_NO is the identifier for each observation. However I want to setup panel data where SCH_NO and STU_NO are identifiers (i think its called stacking in eviews?)

I would like to be able to run panel data regressions as the following:

Y_ij c X1_ij X2_ij where i = student number and j=school.

at the moment I can only run this regression:

Y_i c X1_i X2_i where i = student number.

So my problem is that I cant setup panel data structure in my workspace with an identifier (SCH_NO) that is not a date.

[EViews Glenn]

You can structure by both STU_NO and SCH_NO as the unstructured identifiers.

I must admit that I'm still a bit confused about the structure of your data and how you are going to use the panel identifiers. Are you going to have different coefficients for each student/school pair? Do students move between schools? Is there a time dimension once you've structured by student and school.