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

st: Re: Heisenberg principle


From   Christopher F Baum <baum@bc.edu>
To   statalist@hsphsun2.harvard.edu
Subject   st: Re: Heisenberg principle
Date   Fri, 18 Oct 2002 07:55:42 -0400

--On Friday, October 18, 2002 2:33 -0400 Joseph wrote:

 under what circumstances would time be considered a random effect?
I was going to run this by my friend the theoretical physicist, but...

It is not time (nor the individual, for that matter: state of residence is not a random choice, either) that is random, it is the deviation of this unit's observations from the grand mean that is taken as random. I.e.

y_{i,t} = mu + X_{i,t} b + [theta_i + epsilon_{i,t}]

makes the assumption that the 'effect' of being Ms. I is a random variable
theta_i (rather than a 'fixed effect'). There is nothing wrong with instead having lambda_t as the random effect on y of being in period t. It is the change in the conditional mean of y that is random, not the time period.

Kit
*
* 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