Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: Panel ID and Time Variable


From   "jyotsna puri" <pjyotsna@hotmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   RE: st: Panel ID and Time Variable
Date   Mon, 11 Apr 2005 12:20:48 +0000

Shyamal,
You wouldnt have to re-enter your data to use
averages across nations, as your observational units.
You could just use -collapse- and then do your panel
estimation.

HTH.
-Jyotsna


Jyotsna (Jo) Puri
pjyotsna@hotmail.com
Doctoral Candidate, Department of Agriculture and Resource Economics,
University of Maryland, College Park, MD




----Original Message Follows----
From: "Shyamal Kumar Das" <shyamal@srt.com>
Reply-To: statalist@hsphsun2.harvard.edu
To: statalist@hsphsun2.harvard.edu
Subject: st: Panel ID and Time Variable
Date: Mon, 11 Apr 2005 05:50:05 -0400 (EDT)

Hi:
I am new in statalistserv. Let me describe my problem. I
have a cross-national dataset for 93 countries that I
collected from ICPSR's World Value Surveys. Now, in this
dataset, the unit of analysis is INDIVIDUAL (a total of
approximately 42,000 individuals for 93 countries) with
three waves (years): 1981, 1990, and 1997; when I want to
run Random or Fixed effect models, and put "nations" as
Panel ID, I get results, but when I put "nations" as Panel
ID and "Waves" as Time variable, STATA suggests that there
are repeated time values in the panel. I guess this is
because each individual (unit of analysis) has the same
year. For example, 4000 indivials of France in 1981 have
1981 as time variable. Perhaps, therefore, I do not get any
result out of it. I know that by calculating nations'
averages, and entering these as new variables might solve
the problem. I was loooking for some simpler solution to the
problem so that I do not need to reenter data. I have two
questions in this regard: (1) How can I do panel analysis
using this dataset putting "Nations" and "Waves" as my Panel
ID and Time variable respectively? (2) If I do not put Time
variable, but put Panel ID and run regression, and get
results, is it ok for my panel analysis?
I would appreciate if anyone kindly enlighten me in the
above problem. This will save me at this point.

Shyamal
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index