# Extension: crossed RE's

Adds a "day effect" following Millar (2004, Aust NZ J. Stat, 46, p. 543-554)

The following example was considered by Millar (2004, Aust NZ J. Stat, 46, p. 543-554).
A "day effect" (

where

*v*) was added to the original model formulation, yielding*y*= f

_{ij}_{1,ij}/(1 + exp[-(

*t*-f

_{2})/f

_{3}]) ] + e

_{ij},

f

_{1,ij}= f_{1}+*u*+_{i}*v*_{j}

_{}*u*is a tree-effect and

*v*is a day-effect. This is an example of a model where the random effects

*u*and

*v*are crossed. Millar (2004) used simulated likelihood to evaluate the marginal likelihood.