Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • What to do with categorical variables in SUREG with fixed effects

    Dear all,

    I am currently working with panel data and I want to conduct a SUR regression using sureg for a number of regressions including fixed effects. From other forum answers I gathered that to include fixed effects to the sureg regressions that I can use the xtdata command to demean the data then run -sureg- with the demeaned variables. However, I cannot include binary variables as these are also demeaned in the process. My question is if there is a simple way to include these binary variables after performing xtdata, as only the demeaned variables remain in the dataset. I have been unable to merge this data with the binary variables individually as the year variables cannot be included in xtdata as that also becomes demeaned. The data i have for the binary variables is by country and year, and i cannot merge by country_id as this is not individual to observation but by country, and so the same for each year for each country in the dataset. I apologize if this is a little difficult to follow but I hope someone may have some advice.
    Thank you in advance!

    Nuni B
Working...
X