Predict from a zero-inflated Poisson/NB fit

Usage

# S3 method for class 'zi_count'
predict(
  object,
  newdata = NULL,
  type = c("response", "count", "zero", "intensity"),
  offset = NULL,
  se.fit = FALSE,
  interval = c("none", "confidence"),
  level = 0.95,
  ...
)

Arguments

object: A "zi_count" object.
newdata: Optional data frame of covariate profiles. A fixed-effects unit an equation has not seen, or every row when its column is absent, enters that equation at the reference unit. A zero_fe unit with no positive count gets the structural-zero probability one and one whose every count is positive zero, and an fe unit with no positive count a count mean of zero (see compare_models(), Degrees of freedom).
type: "response" (marginal E(Y)), "count" (equivalently "intensity"; the count component's mean E(Y) of that component), or "zero" (structural-zero probability). "intensity" is accepted as a synonym for "count" for consistency with predict.zi_cpb().
offset: Optional offset (log scale) of the count component for the newdata rows; by default the fit's own offset column is read from newdata (see predict.cpb()).
se.fit, interval, level: Standard errors (se.fit = TRUE) and a level confidence interval (interval = "confidence") of the predicted quantity at the rows of newdata; see predict.cpb().
...: Unused.

Value

A numeric vector. With interval = "confidence", a matrix with columns fit, lwr and upr; with se.fit = TRUE, a list of fit (the vector or the matrix), se.fit and method (each row's interval source). Without se.fit, the source is the attribute method.