\(\newcommand{\B}[1]{ {\bf #1} }\) \(\newcommand{\R}[1]{ {\rm #1} }\)
mixed_ran_likelihood#
View page sourceRandom Effects Likelihood#
Syntax#
v = ran_likelihood.forward (0, theta , u )
ran_likelihood#
is a d_fun representation of the negative log of the random effects likelihood
\[r( \theta , u )
=
v_0 ( \theta , u )
=
- \log [ \B{p} ( y | \theta , u ) \B{p} ( u | \theta ) ]\]
The function \(v_0 ( \theta , u )\) is assumed to be a smooth w.r.t the vector \(( \theta , u )\).
theta#
is a numpy vector with float elements and size
n_fixed
containing a value for the fixed effects.
u#
is a numpy vector with float elements and size
n_random
containing a value for the random effects.
v#
is a numpy vector with float elements and size 1.
None#
The value ran_likelihood = None
corresponds to the random effects likelihood
being constant w.r.t. \(( \theta , u )\).