Extract influence function from model object

Extract i.i.d. decomposition (influence function) from model object

Usage

# Default S3 method
IC(x, bread, id = NULL, ...)

Arguments

x

model object

bread

(optional) Inverse of derivative of mean score function

id

(optional) id/cluster variable

...

additional arguments

Value

Matrix with rows corresponding to observations and columns to parameters, representing the estimated influence function. Attributes include bread (derivative of the estimating equation).

See also

var_ic, iid

Examples

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)  1.005602e-02 -4.173875e-05 -0.01007301
#> x           -4.173875e-05  6.989745e-03  0.00149390
#> z           -1.007301e-02  1.493900e-03  0.02068351