When I use Random effects linear models with and AR(1) disturbance
estimation, I got:
1. dropped constant
2. rho_ar | -.085 (estimated autocorrelation coefficient)
sigma_u | 0
sigma_e | 348.752509
rho_fov | 0 (fraction of variance due to u_i)
I don't know what command Ben has used; -xtreg- will not estimate a
model with an AR(1) correction. Was this output from -xtregar-? If
so, which option was used to define the nature of the AR(1) process?
Per the Statalist FAQ, you should spell out what you said and what
Stata said in response.