Skip to contents

Calculate Summary Statistics

Source: R/calculate_summary_stats.R
calculate_summary_stats.Rd

This function calculates various summary statistics for formatted output of PKNCA::pk.nca().

Usage

calculate_summary_stats(data, input_groups = "ATPTREF")

Arguments

data

A data frame containing results of Non Compartmental Analysis using PKNCA package. Assumes presence of columns: PPORRES, PPSTRES, PPSTRESU, PPTESTCD

input_groups

A character vector specifying the columns to group by. Here. the hierarchical order matters Default is "PPSTRESU".

Value

A data frame with summary statistics for each group and parameter.

Details

The function calculates the following statistics for numeric variables:

  • Geometric mean (geomean)

  • Geometric coefficient of variation (geocv)

  • Arithmetic mean (mean)

  • Standard deviation (sd)

  • Minimum value (min)

  • Maximum value (max)

  • Median value (median)

  • Count of missing values (count.missing)

  • Count (count)

The resulting summary statistics are rounded to three decimal places. If units are different, they are standardized to the group's most frequent first unit.

Examples

data <- data.frame(
  ATPTREF = c(1, 1, 1, 1, 1, 1),
  PPTESTCD = c("A", "A", "B", "B", "C", "C"),
  PPORRES = c(10, 20, 5, 15, NA, 30),
  PPSTRES = c(10, 20, 5, 15, NA, 30),
  PPORRESU = c("mg/L", "mg/L", "ng/mL", "ng/mL", "µg/L", "µg/L"),
  PPSTRESU = c("mg/L", "mg/L", "ng/mL", "ng/mL", "µg/L", "µg/L")
)
calculate_summary_stats(data)
#> # A tibble: 9 × 5
#>   ATPTREF Statistic     `A[mg/L]` `B[ng/mL]` `C[µg/L]`
#>     <dbl> <chr>             <dbl>      <dbl>     <dbl>
#> 1       1 Geomean           14.1        8.66        30
#> 2       1 Geocv             50         81.6         NA
#> 3       1 Mean              15         10           30
#> 4       1 SD                 7.07       7.07        NA
#> 5       1 Min               10          5           30
#> 6       1 Max               20         15           30
#> 7       1 Median            15         10           30
#> 8       1 Count.missing      0          0            1
#> 9       1 Count.total        2          2            2