Desktop productivity for business analysts and programmers

EG: unbalanced panel data regression

Posts: 1

EG: unbalanced panel data regression

As a total newbie in SAS, being unfamiliar with the 'real' SAS programming, I decided to use EG to run a linear regression on a unbalanced panel data set (through Task>Time Series> Regression Analysis of Panel data) consisting of about 180.000 rows.

While doing so however, i get error codes on either the time-series or the cross section since there is plenty of missing data:

"ERROR: There is only one cross section or time series observation"


"ERROR: Not enough observations with non-missing model variables for model statement in cross section GVKEY=001411"

practically, I assume this means that Cross-section items appearing only once in the dataset should be deleted, right? How can I best approach this problem taking into account my lack of programming skills?

Thanks in advance,

New User
Posts: 1

Re: EG: unbalanced panel data regression

Hi NewbieSAS, 


I had the same problem. This is what I did. Open the data set in excel. Select the column with the cross-section ID (in my case it is firm ID number). Use conditional formating icon to highlight all duplicates. Select and delete the non-duplicates. Hope that works. 

Ask a Question
Discussion stats
  • 1 reply
  • 2 in conversation