- ... model
^{1}
- Note that in the FMRI literature
this has previously been referred to as a
*random-effects
model*. Within this paper, however, the separate fixed- and
random-effects contributions to the mixed-effects variance are
considered, thus making a clear distinction between ``random-'' and
``mixed-effects'' important.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

- ... mixed-effects
^{2}
- This naturally includes
fixed-effects tests simply by assuming that the group-level
(co-)variance matrix
is zero.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.