--- "Roy,Suryadipta" <[email protected]> wrote:
> > > > there were initially 1500 cases (no. of year-countries) which
> > > > was subsequently reduced to 100 cases (no. of countries) by
> > > > taking the averages of the variables over 15 years. In this
> > > > situation, will it be alright to report the results I am
> > > > getting for the regression with 1500 observations
--- On Sun, 8 Apr 2007, David Greenberg wrote:
> > > No, it is not.
--- SamL wrote:
> > Why not? I mean, if I understand correctly, the real data is 100
> > observations every year, for 15 years. The averages are throwing
> > away information. Admittedly, one might want to do something to
> > account for the clustering of cases. <snip> But, none of those
> > approaches will just throw away the information that is in the 15
> > year variation in the data, variation that could be useful
--- Tak-wai Chau <[email protected]> wrote:
> As I understand, his original setup is to put "repeated observations"
> of the averages for each country, but not directly putting z1, z2
(the
> time varying variables) into the regression.
This is how I understood the original question as well. So, my and
David's answer means that if you just use the averages, than you have
only a 100 observations. Sam's answer is, it is probably not a good
idea to just use averages, since countries change over time and this
change is informative. I agree with both, however understanding the
alternative models needed is not easy. Judging from the original
question Roy will have a lot of studying ahead of him if he chooses to
take Sam's advise.
Another point is that if you have a 100 countries you almost have the
entire population, in which case significance isn't an issue at all. In
frequentist statistics (virtually all statistics done by Stata) all
uncertainty is due to drawing a sample. It is this uncertainty that is
described by standard errors, confidence intervals, tests, etc. If you
do not draw a sample, but study the entire population, there is no
uncertainty left. A parameter just is the parameter, you may think it
is large or small, and describe it as such, but it does no longer make
any sense to perform test on it. So if you are willing to declare those
100 countries your population, than it doesn't matter that the
estimates are no longer significant.
Hope this helps,
Maarten
-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
___________________________________________________________
Inbox full of unwanted email? Get leading protection and 1GB storage with All New Yahoo! Mail. http://uk.docs.yahoo.com/nowyoucan.html
*
* 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/
