rlmerMod object.Extends implicitIF by adding the \(\sigma\)-DAS and
\(\theta\)-DAS scoring equations to the implicit-derivative linear
system. The Jacobian of the full score wrt \((\beta, u, \log
\sigma, \theta)\) is computed via numDeriv::jacobian
with Richardson extrapolation, reusing the converged DAS scales on
the non-\(\theta\) columns (they depend on \(\theta\) only); the
Jacobian wrt the response is closed-form. Only DASvar and
DAStau methods are supported (DAStau additionally requires
block sizes \(\le 2\)).
implicitIF_full(fit, eps = 1e-06, use.cache = TRUE)A list with components IF_beta (\(p \times n\)),
IF_u (\(q \times n\)), IF_sigma (\(1 \times n\)),
IF_theta (\(L \times n\)), and the model-based delta-method
covariance vcov_model_delta.
An rlmerMod object.
Numerical step for numDeriv::jacobian (default
1e-6).
If TRUE (default), return a result cached on
the fit when one is stored for the current \(\theta\), and cache
the result otherwise.
This function is the engine behind cooks.distance
/ influence; users who only want
Cook's distance should call those S3 methods.
The result is cached on the fit (keyed by \(\theta\)) when
use.cache = TRUE, so repeated calls -- and the consumers
cooks.distance, the sandwich
vcov, and the Satterthwaite df of summary /
anova / emmeans / confint -- share a single
computation.
implicitIF,
cooks.distance