Standardise Missing Value Indicators to NA

Converts custom "missing" representations such as "-", "Missing", "N/A", "none", "null", "", or any user-defined string/number into proper NA values so that the rest of the pipeline can detect and handle them correctly.

Usage

standardize_na(
  data,
  na_strings = character(0),
  na_numbers = NULL,
  cols = NULL,
  trim = TRUE,
  verbose = TRUE
)

Arguments

data

A data.frame.

na_strings

Character vector. Additional strings to treat as NA. These are combined with the built-in defaults. Set to character(0) to use only the defaults.

na_numbers

Numeric vector. Numeric sentinel values to treat as NA, e.g. -999, 9999, 0. Default NULL (none).

cols

Character vector. Column names to process. NULL (default) processes all columns.

trim

Logical. If TRUE (default), trim leading/trailing whitespace from character columns before matching.

verbose

Logical (default TRUE).

Value

The data.frame with all matched values replaced by NA.

Examples

df <- data.frame(
  date  = c("2024-01-01", "2024-01-02", "2024-01-03"),
  temp  = c(25.1, -999, 26.2),
  humid = c("80", "-", "Missing"),
  wind  = c("N/A", "12", "")
)

# Use defaults + custom
clean <- standardize_na(df,
  na_strings = c("N/A"),
  na_numbers = c(-999)
)
#> 
#> ============================================================
#>   STANDARDISE MISSING VALUES
#> ============================================================
#> 
#>   Built-in NA strings : 31 patterns
#>   User NA strings     : N/A
#>   NA numbers          : -999
#>   Cells converted     : 5
#> 
#> ------------------------------------------------------------
#>   Conversions per Column
#> ------------------------------------------------------------
#>  column converted
#>    temp         1
#>   humid         2
#>    wind         2
#> 
clean
#>         date temp humid wind
#> 1 2024-01-01 25.1    80 <NA>
#> 2 2024-01-02   NA  <NA>   12
#> 3 2024-01-03 26.2  <NA> <NA>