Re: st: conception confusion - "fixed effects" and time effect on data with time factor

[Maarten Buis <maartenlbuis@gmail.com>]

On Tue, Oct 18, 2011 at 3:15 AM, House Wang wrote:
> Yes, the data is like survey done across 30 years, each time with same
> measures, but on different persons.
> How could I understand time effect on this kind of data? Is there fixed
> effects on them? and how to test them?

You just have cross-sectional data, so forget the comparison with
panel data. Time is just a regular variable like any other, and it
measures the period at which the interview took place. This can be
relevant when you want to analyze questions that are explicitly or
implicitly referring to "now". However, you need to decide whether
period is the right time axis. A common alternative is cohort, i.e.
year of birth. Also note that you cannot include age, period and
cohort in your model as they will be perfectly collinear: cohort =
period - age.

Hope this helps,
Maarten


--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany


http://www.maartenbuis.nl
--------------------------
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/