Skip to contents

Generate rowwise standard deviation

Source: R/gen_rowsd.R
gen_rowsd.Rd

This function returns the rowwise standard deviation in a data frame.

Usage

gen_rowsd(data, cols, na.rm = TRUE)

Arguments

data

A data frame.

cols

<tidy-select> Columns to search across. Non-numeric columns are ignored.

na.rm

Logical vector of length one passed onto stats::sd().

Value

A double vector of the rowwise standard deviation.

Details

Parallelization is supported via purrr::in_parallel().

If na.rm is TRUE (the default), missing values are removed. If na.rm is FALSE, the presence of an NA or NaN in any row will return NA for that row. As per the documentation for stats::sd(), the standard deviation of a length one vector is NA.

Examples

library(dplyr, warn.conflicts = FALSE)
a <- tibble(
  x = c(1, NA, 2),
  y = c(NA, 3, NA),
  z = c(4, NA, 5),
  aa = c("a", "b", "c"),
  bb = rep(NA, 3),
  cc = rep(NaN, 3)
)
gen_rowsd(a)
#> [1] 2.12132      NA 2.12132
gen_rowsd(a, all_of(letters[25:26]))
#> [1] NA NA NA
b <- a %>% mutate(q = gen_rowsd(.), r = gen_rowsd(., na.rm = FALSE))
b
#> # A tibble: 3 × 8
#>       x     y     z aa    bb       cc     q     r
#>   <dbl> <dbl> <dbl> <chr> <lgl> <dbl> <dbl> <dbl>
#> 1     1    NA     4 a     NA      NaN  2.12    NA
#> 2    NA     3    NA b     NA      NaN NA       NA
#> 3     2    NA     5 c     NA      NaN  2.12    NA