model.frame: Model.frame methods in bayesnec.

In bayesnec: A Bayesian No-Effect- Concentration (NEC) Algorithm

Model.frame methods in bayesnec.

Description

Retrieve data.frame used to fit models via bnec, or directly from a bayesnecformula. formula should be of class bayesnecfit, bayesmanecfit or bayesnecformula.

Usage

## S3 method for class 'bayesnecfit'
model.frame(formula, ...)

## S3 method for class 'bayesmanecfit'
model.frame(formula, model, ...)

## S3 method for class 'bayesnecformula'
model.frame(formula, data, ...)

Arguments

formula	An model object of class bayesnecfit, bayesmanecfit, or a formula of class bayesnecformula.
...	Unused if formula is a bayesnecfit or a bayesmanecfit. Else, if formula is a bayesnecformula, additional arguments to be passed to check_formula.
model	A valid model string.
data	A data.frame containing the variables specified in formula.

Details

If formula is a bayesnecformula and it contains transformations to variables x and y, these are evaluated and returned as part of the data.frame.

Value

If formula is a bayesnecfit or a bayesmanecfit, a data.frame containing the data used to fit the model.

If, instead, formula is a bayesnecformula, a data.frame with additional attributes detailing the population-level variables (attribute "bnec_pop") (response y, predictor x, and, if binomial a formula, trials) and, if applicable, the group-level variables (attribute "bnec_group").

Examples

library(bayesnec)
# if input is of class `bayesnecfit` or `bayesmanecfit`
model.frame(manec_example, model = "nec4param")
nec4param <- pull_out(manec_example, "nec4param")
model.frame(nec4param)
# if input is of class `bayesnecformula`
nec3param <- function(beta, nec, top, x) {
  top * exp(-exp(beta) * (x - nec) *
    ifelse(x - nec < 0, 0, 1))
}

data <- data.frame(x = seq(1, 20, length.out = 10), tr = 100, wght = c(1, 2),
                   group_1 = sample(c("a", "b"), 10, replace = TRUE),
                   group_2 = sample(c("c", "d"), 10, replace = TRUE))
data$y <- nec3param(beta = -0.2, nec = 4, top = 100, data$x)

f_1 <- y ~ crf(x, "nec3param")
f_2 <- "y | trials(tr) ~ crf(sqrt(x), \"nec3param\")"
f_3 <- y | trials(tr) ~ crf(x, "nec3param") + ogl(group_1) + pgl(group_2)
f_4 <- y | trials(tr) ~ crf(x, "nec3param") + (nec + top | group_1)

m_1 <- model.frame(bnf(f_1), data)
attr(m_1, "bnec_pop")
model.frame(bnf(f_2), data)
m_3 <- model.frame(bnf(f_3), data)
attr(m_3, "bnec_group")
model.frame(bnf(f_4), data)

bayesnec documentation built on Sept. 26, 2023, 9:06 a.m.