Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: RE: Areg, absorb

From   Nick Cox <>
Subject   Re: st: RE: Areg, absorb
Date   Mon, 11 Apr 2011 14:52:41 +0100

So identifier is "country1 country2" and time variable "year"? Well,
then -group(country1 country2 year)- will define a distinct (and
unique) value for each observation, which is far too many dummies.


On Mon, Apr 11, 2011 at 2:48 PM, emanuele mazzini
<> wrote:

> Exactly country1 country2 and year.
> 2011/4/11 Nick Cox <>:
>> Everything depends on whether you want dummies for _combinations_ of
>> -country1-, -country2- and -year-.
>> What precisely defines a panel?
>> Nick
>> On Mon, Apr 11, 2011 at 1:30 PM, emanuele mazzini
>> <> wrote:
>>> I tried that but doesn't really work. Curiously, Stata reports
>>> insufficient observations. That's kind of strange, however.
>>> 2011/4/11 Jan Bryla <>:
>>>> Would it not be possible for you to first generate group id's based on country1 country2 and year, and then use -areg- to control for each group? I think that would capture the desired effects.
>>>> In that case it would be something like:
>>>> Egen id=group(country1 country2 year)
>>>> Areg ..., abs(id)
>>>> /Jan Bryla
>>>> -----Original Message-----
>>>> From: [] On Behalf Of emanuele mazzini
>>>> Sent: 11. april 2011 13:33
>>>> To:
>>>> Subject: st: Areg, absorb
>>>> Dear all Stata users,
>>>> I am trying to work out a regression with panel data in which I have
>>>> to control for country1, country2 and year.
>>>> I tried to ran a regression after creating a dummy for all three of
>>>> them and obviously in the result Stata reports the estimates of all
>>>> the dummies I created which I do not want to show in the output of the
>>>> regression. I tried to get the same result with the command: areg...,
>>>> absorb(varname)  but the point is that areg does not allow me to
>>>> include more than one variable, and in this case I would need three.
>>>> Hence, my brief question is: is there a way to include in the
>>>> regression the fixed effects for country1, country2 and year without
>>>> Stata showing all the coefficients of the dummies I create?

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index