Zero-inflated Negative Binomial GAM

Source: R/EM_gam.R

Fit a zero-inflated Negative Binomial Generalized Additive Model using the EM Algorithm

zinbgam(
  mu.formula,
  pi.formula,
  data,
  knots = NULL,
  mu = NULL,
  pi = NULL,
  theta = 1,
  gamma.pi = 1,
  gamma.mu = 1,
  select = FALSE,
  method = "GCV.Cp",
  min.em = 5,
  max.em = 50,
  tol = 0.01
)

Arguments

mu.formula

formula for the count model

pi.formula

formula for the binary model

data

a data frame or list containing the model response variable and covariates required by the formula.

knots

an optional list of knot values to be used for basis construction.

mu

initial mu vector

pi

intial pi vector

theta

initial theta value

gamma.pi

binary model gamma

gamma.mu

count model gamma

select

include model selection penalty

method

method for selecting smoothing parameters

min.em

minimum number of EM iterations

max.em

maximum number of EM iterations

tol

tolerance (default=1.0E-2)