Extract i.i.d. decomposition (influence function) from model object
IC(x,...)
# S3 method for default
IC(x, bread, id=NULL, folds=0, maxsize=(folds>0)*1e6,...)
model object
additional arguments
(optional) id/cluster variable
(optional) Inverse of derivative of mean score function
(optional) Calculate aggregated iid decomposition (0:=disabled)
(optional) Data is split in groups of size up to 'maxsize' (0:=disabled)
m <- lvm(y~x+z)
distribution(m, ~y+z) <- binomial.lvm("logit")
d <- sim(m,1e3)
g <- glm(y~x+z,data=d,family=binomial)
var_ic(IC(g))
#> (Intercept) x z
#> (Intercept) 1.000879e-02 -1.825991e-05 -0.010032817
#> x -1.825991e-05 8.222422e-03 0.003055657
#> z -1.003282e-02 3.055657e-03 0.022557940