Skip to contents

Select Appropriate Confidence Interval Method for a drc Model

Source: R/ED_robust.R
get_ed_interval.Rd

This function determines the recommended confidence interval calculation method ('type' argument in drc::ED) based on the model family of a 'drc' object.

Usage

get_ed_interval(
  model,
  small_n = TRUE,
  fls_pattern = "^LL|^LN|^BC|^CRS",
  verbose = FALSE
)

Arguments

model

A drc model object or a character string specifying the model name (e.g., "LL.4").

small_n

A logical value. If TRUE, the t-distribution-based Fieller's method ("tfls") is used for small samples for applicable models. If FALSE, the normal-distribution-based method ("fls") is used. Defaults to TRUE.

fls_pattern

A regular expression character string. This pattern is used to identify model families for which the "fls" or "tfls" method is appropriate. The default covers standard log-logistic, log-normal, Brain-Cousens, and Cedergreen-Ritz-Streibig models.

verbose

A logical value. If TRUE, a message is printed when the function resorts to its default choice because the model type was not explicitly matched. Defaults to TRUE.

Value

A character string: "tfls", "fls", or "delta", representing the recommended interval type for use in drc::ED().

Author

Hannes Reinwald

Examples

ryegrass_model <- drm(rootl ~ conc, data = ryegrass, fct = LL.4())
drc:::get_ed_interval(ryegrass_model)
#> [1] "tfls"
drc:::get_ed_interval("LL.4")
#> [1] "tfls"
drc:::get_ed_interval("W1.4")
#> [1] "delta"