How to turn off scientific notation for estat ic/est stats to avoid identical AIC results?

[Signe Svallfors]

Hi,
I'm in need of advice of how to turn off scientific notation for estat ic/est stats. I have changed the format for the regressions so that scientific notation is not used in those, but I haven't been able to do it for the subsequent commands.

Likely because of the scientific notation, I get identical AIC results for multiple models. Or -- if someone has another idea of why I might get the same AIC results for different models, I'd be very grateful for any advice.

My code looks something like this:
xtreg y x1 x2 ..., , fe vce(cl var)
est stats --OR-- est stats

Best,
Signe

[Andrew Musau]

Scientific notation is used to display either very large or very small numbers. This value will be stored in a matrix -r(S)- which you can retrieve.

webuse grunfeld, clear
qui xtreg invest mvalue kstock i.time, fe
est stats
mat l r(S)
di %17.9f r(S)[1, 5]

Res.:

. est stats

Akaike's information criterion and Bayesian information criterion

-----------------------------------------------------------------------------
       Model |          N   ll(null)  ll(model)      df        AIC        BIC
-------------+---------------------------------------------------------------
           . |        200  -1216.349  -1056.132      22   2156.264   2228.827
-----------------------------------------------------------------------------
Note: BIC uses N = number of observations. See [R] BIC note.

.
. mat l r(S)

r(S)[1,6]
            N         ll0          ll          df         AIC         BIC
.         200  -1216.3487  -1056.1322          22   2156.2645   2228.8275

.
. di %17.9f r(S)[1, 5]
   2156.264484287

[Signe Svallfors]

Big thanks, Andrew! This solution worked perfectly.

Best,
Signe