Learn R Programming

LEGIT (version 1.4.1)

backward_step: Internal function that does the backward step for the stepwise IM function. #@param empty_start_dataset If TRUE, the initial dataset is empty. #@param fit Current best fit. #@param ... Same parameters as in the stepwise function.

Description

Internal function that does the backward step for the stepwise IM function. #@param empty_start_dataset If TRUE, the initial dataset is empty. #@param fit Current best fit. #@param ... Same parameters as in the stepwise function.

Usage

backward_step(
  fit,
  data,
  formula,
  interactive_mode = FALSE,
  genes_current = NULL,
  env_current = NULL,
  genes_dropped = NULL,
  env_dropped = NULL,
  search = "genes",
  search_criterion = "AIC",
  p_threshold = 0.2,
  exclude_worse_AIC = TRUE,
  max_steps = 100,
  cv_iter = 5,
  cv_folds = 10,
  folds = NULL,
  Huber_p = 1.345,
  classification = FALSE,
  start_genes = NULL,
  start_env = NULL,
  eps = 0.01,
  maxiter = 100,
  family = gaussian,
  ylim = NULL,
  seed = NULL,
  print = TRUE,
  test_only = FALSE
)

Arguments

Value

Returns fit, start_genes, start_env and genes_current, genes_dropped if search="genes" or env_current and env_dropped if search="env".