st: Re: Heisenberg principle

[Christopher F Baum <baum@bc.edu>]

--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/