Create multiple new variables that are calculated from existing variables

user_9462675 0
Viewed 63

I'm trying to figure out how to create multiple new variables that are calculated using variables currently in my dataset

Here's some example data

library(tidyverse)

df <- data.frame(
 a1 = rnorm(100),
 a2 = rnorm(100),
 b1 = rnorm(100),
 b2 = rnorm(100),
 c1 = rnorm(100),
 c2 = rnorm(100)
)

Essentially, I want to create a new variable for each a, b, c pair that divides a1 by a2, b1 by b2, etc. For example:

df <- df %>%
  mutate(a3 = a1/a2)

The variables in my dataset don't follow these naming conventions, so I feel like I need to assign names to vectors:

numerators <- c('a1', 'b1', 'c1')
denominators <- c('a2', 'b2', 'c2')

And then creating a new variable would follow the basic convention

mutate(newvars = numerators/denominators)

But I'm stuck as to how to actually do this. Any help would be much appreciated -- thanks!

purrr dplyr iteration r
1 Answers

Here is an option to split the dataset into a list of data.frame based on the column name pattern, then reduce it by dividing elementwise on each pair of columns in each of the dataset and bind with the original dataset

library(tidyverse)
df %>% 
   split.default(sub("\\d+", "", names(.))) %>% 
   map_df(reduce, `/`) %>% 
   rename_all(~paste0(., 3)) %>% 
   bind_cols(df, .)
user_37322710
Related