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

Re: st: Panel ID and Time Variable

Subject   Re: st: Panel ID and Time Variable
Date   Mon, 11 Apr 2005 13:46:25 +0200


I guess, you must first -tsset- your data, i.e.

. tsset nations waves

and then run your regressions.

Hope this helps.


Zitat von Shyamal Kumar Das <>:

> 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:
> *
> *
> *

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index