R: Tidy variance structure for the 'nlme' package.

tidy.varFunc {broom.mixed}

R Documentation

Tidy variance structure for the `nlme` package.

Description

Returns a tibble with the following columns:

group: type of varFunc, along with the right hand side of the formula in parentheses e.g. "varExp(age | Sex)".
term: terms included in the formula of the variance model, specifically the names of the coefficients. If the value is fixed, it will be appended with " ; fixed".
estimate: estimated coefficient
estimated: This column is only included if some parameters are fixed. TRUE if the parameter is estimated and FALSE if the parameter is fixed.

Usage

## S3 method for class 'varFunc'
tidy(x, ...)

## S3 method for class 'varComb'
tidy(x, ...)

Arguments

`x`	An object of class `varFunc`, such as those used as the `weights` argument from the `nlme` package
`...`	Ignored

Value

If the varFunc is uninitialized or has no parameters, the function will return an empty tibble. Otherwise, it will return a tibble with names described in the details section.

Examples

## Not run: 
if (require("nlme")) {
ChickWeight_arbitrary_group <- datasets::ChickWeight
ChickWeight_arbitrary_group$group_arb_n <-
  1 + (
    as.integer(ChickWeight_arbitrary_group$Chick) >
    median(as.integer(ChickWeight_arbitrary_group$Chick))
  )
ChickWeight_arbitrary_group$group_arb <- c("low", "high")[ChickWeight_arbitrary_group$group_arb_n]

fit_with_fixed <-
  lme(
    weight ~ Diet * Time,
    random = ~Time | Chick,
    data =ChickWeight_arbitrary_group,
    weights=varIdent(fixed=c("low"=5), form=~1|group_arb)
  )
# Show all parameters
tidy(fit_with_fixed)
# Exclude fixed parameters
tidy(fit_with_fixed) %>%
  filter(across(any_of("estimated"), ~.x))
}

## End(Not run)