st: RE: random quadratic effect without random linear effect?

[Timothy Mak <tshmak@hku.hk>]

Hi Jordan, 

Why do you say it may make theoretical sense to include only a random quadratic effect? 

I think to have a quadratic term that is a random effect is highly unusual already. I cannot think of why you would then want it to be the only random effect. If you want to have nonlinear trends, splines (see -bspline- on SSC) and fractional polynomials (-fracpoly-) are the usual considerations, although I think these are rarely considered in a hierarchical model (i.e. we don't hear very often of "random splines" and "random fractional polynomials") although in theory it can be done by having all of the terms as random effects. Again, I can think of no theoretical reason why only one of the terms should be made random. 

Tim

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Jordan Silberman
Sent: 20 June 2013 06:07
To: statalist@hsphsun2.harvard.edu
Subject: st: random quadratic effect without random linear effect?

Hi Stata Folks,

For a model run in xtmixed, we all know that if you include a fixed
quadratic effect of predictor A, then you typically should include the
lower order fixed linear effect of A as well.

Does an analogous rule apply to random effects? I think that it may
make theoretical sense to include only a random quadratic effect,
without the corresponding random linear effect. Can anyone tell me
whether or not this would be "kosher?"

Thanks,
Jordan
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/

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