Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

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

Re: st: WG: xtreg command and two-way fixed o random effect model

From   David Jacobs <>
Subject   Re: st: WG: xtreg command and two-way fixed o random effect model
Date   Sun, 09 May 2010 15:44:20 -0400

To get two-way fixed effects in official Stata, one must add the year dummies. This is easily done in Stata 11 with an term like: -i.year- assuming that "year' is the variable name for your time variable. In Stata 8, a separate command like: -tabulate year, gen(yrdum)- will provide permanent dummies for time with "yrdum" being the stub name for each of the separate dummy variables for each of the periods in your data set.

Although the econometrically sophisticated probably can provide work-arrounds, it is not possible to automatically include random effects for time in the -xtreg- routines. Others, however, may know of user-written programs that do this.

Both year fixed-effects and random-effects can are readily provided in the econometrics package called EViews, and probably (although I can't remeber for sure) in another econometrics package called Limdep.

David Jacobs

At 07:38 AM 5/9/2010, you wrote:
Dear all,

I am working with a balanced panel data set and want to analyze the group
and time effects.

Is there any possibility to use the xtreg command in combination with a
two-way fixed or/and random effect model ?   Maybe at a more updated Stata
Version ? I am currently working  with the Stata 8.0 version?..

Background to my question:
At the moment I apply the following regression model and would like to cross
check these results with the ones obtained from the fixed effect model. (As
is it possible in the case of a one-way model)

 xi: reg y i.idfirm i.idyear x1 x2 x3

Thanks for your inputs, which are highly appreciated.


*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2016 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index