cvms (version 1.7.0)

combine_predictors: Generate model formulas by combining predictors

Description

lifecycle::badge("maturing")

Create model formulas with every combination of your fixed effects, along with the dependent variable and random effects. 259,358 formulas have been precomputed with two- and three-way interactions for up to 8 fixed effects, with up to 5 included effects per formula. Uses the + and * operators, so lower order interactions are automatically included.

Usage

combine_predictors(
  dependent,
  fixed_effects,
  random_effects = NULL,
  max_fixed_effects = 5,
  max_interaction_size = 3,
  max_effect_frequency = NULL
)

Value

list of model formulas.

E.g.:

c("y ~ x1 + (1|z)", "y ~ x2 + (1|z)", "y ~ x1 + x2 + (1|z)", "y ~ x1 * x2 + (1|z)").

Arguments

dependent

Name of dependent variable. (Character)

fixed_effects

list of fixed effects. (Character)

Max. limit of 8 effects when interactions are included!

A fixed effect name cannot contain: white spaces, "*" or "+".

Effects in sublists will be interchanged. This can be useful, when we have multiple versions of a predictor (e.g. x1 and log(x1)) that we do not wish to have in the same formula.

Example of interchangeable effects:

list( list( "x1", "log_x1" ), "x2", "x3" )

random_effects

The random effects structure. (Character)

Is appended to the model formulas.

max_fixed_effects

Maximum number of fixed effects in a model formula. (Integer)

Max. limit of 5 when interactions are included!

max_interaction_size

Maximum number of effects in an interaction. (Integer)

Max. limit of 3.

Use this to limit the n-way interactions allowed. 0 or 1 excludes interactions all together.

A model formula can contain multiple interactions.

max_effect_frequency

Maximum number of times an effect is included in a formula string.

Author

Ludvig Renbo Olsen,

Examples

# Attach packages
library(cvms)

# Create effect names
dependent <- "y"
fixed_effects <- c("a", "b", "c")
random_effects <- "(1|e)"
# \donttest{
# Create model formulas
combine_predictors(
  dependent, fixed_effects,
  random_effects
)
# }
# Create effect names with interchangeable effects in sublists
fixed_effects <- list("a", list("b", "log_b"), "c")
# \donttest{
# Create model formulas
combine_predictors(
  dependent, fixed_effects,
  random_effects
)
# }