consolidate_categories_other: Consolidate categorical variable's rarely observed levels...

In abrahamalex13/doPrepExplore: Standard Data Preparation/Exploration Workflows

Description

Two options, controlled by construct_indic_other:

• (FALSE) Construct modified version of original categorical variable. Overwrite some level values with 'other', according to minimum frequency threshold. Modified variable result takes '_consol' suffix.

• (TRUE) Construct binary indicator variable strictly to convey, 'is other', according to minimum frequency threshold.

Usage

consolidate_categories_other(
  df,
  varname_cat,
  threshold,
  value_consol = "OTHER",
  construct_indic_other = FALSE
)

Arguments

df	A data frame, with categorical variable whose rarely observed levels should be consolidated.
varname_cat	Character name of categorical variable.
threshold	Minimum share of observations with categorical level v, below which, v enters 'other'.
value_consol	Value which constitutes the newly consolidated 'other' category.
construct_indic_other	Does construct binary indicator variable strictly to convey, 'is_other'.

Value

Modified version of input data frame.

abrahamalex13/doPrepExplore documentation built on Jan. 27, 2021, 4:30 a.m.