FD Model

[Cla Tuor]

Hello all,

I am estimating a first differences model of the following form for 16 countries and 12 time periods: ΔY_it = ß₀ + ß₁ΔX_it + ø_t + λ_i + ɛ_it
The available panel data is already in differences, i.e., I have data on ∆Y_it and ∆X_it only (Y_it and X_it are not available).
∆Y_it and ∆X_it are saved as diff_y and diff_x in my Excel dataset; hence I use the following code to estimate the model:

xtset country year
xi: xtreg diff_y diff_x, fe vce(cluster country)

Is that OK like that, or do I need to tell Stata somehow that I am estimating a first-differences model?

Thank you!

[Jared Greathouse]

Uhhhh..... take off the xi:

and you should be fine. FE models don't (I think) care if your outcome is already in a difference format.

[Cla Tuor]

Thank you! I forgot to add time dummies (i.year) in my code. That's why I added xi.