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: unconditional fixed effect logit

From   "Noh, Grimm" <[email protected]>
To   "[email protected]" <[email protected]>
Subject   RE: st: unconditional fixed effect logit
Date   Sun, 30 Jan 2011 16:15:37 +0000

Sorry, but I didn't quiet understand.
I want to know how I can run unconditional fixed effect logit, controlling for unobserved firm effects,
when my data structure is firm-country-year, not firm-year.

I didn't quiet get how egen helps me do that.
More explanation please. Thank you.


From: [email protected] [[email protected]] on behalf of Kit Baum [[email protected]]
Sent: Sunday, January 30, 2011 8:40 AM
To: [email protected]
Subject: re: st: unconditional fixed effect logit

The unit of analysis for my data is firm-country-year and my dependent variable is 0 or 1. If firm A entered country X in year 2000, the value for dependent variable is 1 for firmA - CountryX - yr2000 combination.

when I declare data before logit, if I try to group the data by "firms" to control for each firm's fixed effect, the stata won't run this because for each "firm", the data varies not only by years but also by country.

help egen. Look for the group() function.

Kit Baum
[email protected]

*   For searches and help try:
*   For searches and help try:

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