Skip to contents

Write covariate imputation code

Source: R/write_mutate_code.R
write_covariate_imputation_code.Rd

Generates tidyverse-style `mutate()` code to impute missing values for selected variables. Numeric variables are imputed with the median, while factor or character variables are imputed with the mode (user-defined). In addition to the imputed variable (with suffix `_nona`), a missingness dummy variable (with suffix `_missing`) is created for each input.

Usage

write_covariate_imputation_code(data, ..., include_missingness_dummies = TRUE)

Arguments

data

A data frame or tibble.

...

Columns to generate imputation code for. You can specify them unquoted (e.g., `age`, `income`) or using selection helpers such as [dplyr::all_of()] or [tidyselect::starts_with()]. If left empty, all `"X_"` columns are used.

include_missingness_dummies

Logical. Should missingness dummy variables be included in the generated code? Defaults to TRUE.

Value

Invisibly returns the generated code as a single string.

Details

This function prints imputation code to the console that you can copy-paste into your analysis script. It does not perform the imputation itself.

See also

Other code generators: write_attrition_check_code(), write_balance_check_code(), write_outcome_missingness_dummies_code()

Examples

# Example data with missingness
dat <- data.frame(
  X_factor_variable = factor(rep(c("A", "B", NA), c(4, 5, 1))),
  X_numeric_variable = c(100, 200, NA, 400, 500, NA, 700, 800, 900, NA)
)

# Generate imputation code
write_covariate_imputation_code(dat, X_factor_variable, X_numeric_variable)
#> dat <-
#>   dat |>
#>   mutate(
#> X_factor_variable_nona = replace_na(X_factor_variable, stat_mode(X_factor_variable)),
#> X_factor_variable_missing = if_else(is.na(X_factor_variable), 1, 0),
#> X_numeric_variable_nona = replace_na(X_numeric_variable, median(X_numeric_variable, na.rm = TRUE)),
#> X_numeric_variable_missing = if_else(is.na(X_numeric_variable), 1, 0)
#>   ) 

# Or default to all "X_" columns
write_covariate_imputation_code(dat)
#> dat <-
#>   dat |>
#>   mutate(
#> X_factor_variable_nona = replace_na(X_factor_variable, stat_mode(X_factor_variable)),
#> X_factor_variable_missing = if_else(is.na(X_factor_variable), 1, 0),
#> X_numeric_variable_nona = replace_na(X_numeric_variable, median(X_numeric_variable, na.rm = TRUE)),
#> X_numeric_variable_missing = if_else(is.na(X_numeric_variable), 1, 0)
#>   ) 

# Or use tidyselect helpers
vars <- c("X_factor_variable", "X_numeric_variable")
write_covariate_imputation_code(dat, dplyr::all_of(vars))
#> dat <-
#>   dat |>
#>   mutate(
#> X_factor_variable_nona = replace_na(X_factor_variable, stat_mode(X_factor_variable)),
#> X_factor_variable_missing = if_else(is.na(X_factor_variable), 1, 0),
#> X_numeric_variable_nona = replace_na(X_numeric_variable, median(X_numeric_variable, na.rm = TRUE)),
#> X_numeric_variable_missing = if_else(is.na(X_numeric_variable), 1, 0)
#>   )