Skip to contents

Update the parameters of a pcens object

Source: R/pcens.R
update.pcens.Rd

Replaces the delay distribution parameters, and optionally the primary event distribution arguments, of an existing pcens object. The delay and primary distribution functions, the primary event CDF and the class are kept as they are. The updated object therefore dispatches to the same pcens_cdf() method without looking up names or rebuilding the class. This is cheaper than calling new_pcens() again when one distribution is evaluated for many parameter sets, for example posterior draws.

Usage

# S3 method for class 'pcens'
update(object, ..., primary_args = NULL)

Arguments

object

A pcens object as created by new_pcens().

...

Named delay distribution parameters. Each one replaces the entry of the same name in object$args, or is added if not present. Parameters that are not given keep their current values.

primary_args

Optional named list of primary event distribution arguments. These are merged into object$primary_args in the same way as ... is merged into object$args. Defaults to NULL, which leaves the primary event distribution arguments unchanged.

Value

A pcens object with the same class as object and updated args, primary_args and dprimary_args fields. See new_pcens() for the fields of a pcens object.

Details

Parameters are merged rather than replaced as a whole, so update(object, scale = 3) changes scale and keeps all other parameters. Parameters cannot be removed; use new_pcens() for that.

A name in ... that is not already in object$args must be an argument of object$pdist other than its first, unless pdist takes .... Otherwise an error is raised. Names in primary_args are not checked against dprimary.

See also

Low level primary event censored distribution objects and methods new_pcens(), pcens_cdf(), pcens_cdf.default(), pcens_cdf.pcens_pdiscretehazard(), pcens_cdf.pcens_pdiscretestep(), pcens_cdf.pcens_pgamma_dunif(), pcens_cdf.pcens_pgengamma.orig_dunif(), pcens_cdf.pcens_pgengamma_dunif(), pcens_cdf.pcens_plnorm_dunif(), pcens_cdf.pcens_pweibull_dunif(), pcens_pmf(), pcens_pmf.default(), pcens_quantile(), pcens_quantile.default()

Examples

obj <- new_pcens(
  pdist = pgamma, dprimary = dunif,
  primary_args = list(min = 0, max = 1),
  shape = 1, scale = 1
)
obj <- update(obj, shape = 2, scale = 3)
pcens_cdf(obj, q = c(1, 5, 10), pwindow = 1)
#> [1] 0.01571917 0.44166025 0.82397787

# Update the primary event distribution arguments
obj <- new_pcens(
  pdist = pgamma, dprimary = dexpgrowth,
  primary_args = list(r = 0.2),
  shape = 2, scale = 3
)
obj <- update(obj, primary_args = list(r = 0.5))
pcens_pmf(obj, x = 0:5, pwindow = 1)
#> [1] 0.01385490 0.07361513 0.11112116 0.12144831 0.11699152 0.10530254