Problems using asdoc with fixed effects panel estimation

[Fernando Lopez]

Hello, im having troubles exporting my regression estimates without the confidence intervalls with the asdoc command and the noci option.

asdoc xtreg lnscope lncsrstrat lnsocial lnenvironment lncg lncprw lndebtw lncapw lnpbw i.year i.industry, re noci

This is the command i enter and stata tells me the subscript is invalid and a function returned error, anybody know how to fix this problem? Syntax as according to the guide as far as i know

[George Ford]

I am able to replicate your problem. asdoc does not like the noci option (the coeflegend works, but won't give you the t-stats). Maybe A Shah will join the conversation.

Also, does i.industry include the fixed effects in your re model?

[George Ford]

help asdoc

does indicate that noci is an option.

it does work with reghdfe (but not with xtreg fe)

interesting.

(*added: I take that back, it will run but still reports the ci)

[Fernando Lopez]

George Ford yes those are Industry dummies, i mispelled the modell i use is an random effects model as you already figured out correctly by the use of those dummies. I think the problem has to do with the model specification command and the noci command confusing asdoc, since theyre both behind the comma.

[Attaullah Shah]

I shall fix it in the next update, cannot promise time. In the meanwhile, I checked the wide regression of asdocx, it works correctly.

*Load example data
sysuse auto, clear
‎
* wide regression, suppressing the confidence interval
asdocx reg price mpg trunk i.rep78 headroom weight foreign, replace noci

[George Ford]

I think you're better offer using -mixed- if you want to include the FE in an RE model.