Subset larger Dataframe into smaller ones in R?-CodePudding

I have a larger dataframe from which I would like to split up based on 2 columns and a changing 3rd column.

Am on mobile so hard to give a reproducible example so I will try my best to describe.

I have a large Dataframe with 10 columns, the first 2 being ID and Year.

I would like to have smaller ones where the 3rd column will be each of the remaining 8.

So a total of 8 smaller dataframes

I have tried:

newDF1<-select(BIGdf, c("ID", "Year", "3rdVariable"))

newDF2<-select(BIGdf, c("ID", "Year", "4thVariable"))

And achieve the result but is there a way I don't have to write out each individual variable. Sorry for the poor formatting any help would be appreciated.

CodePudding user response：

It is usually bad practice to split up data which belongs together. However, you can automatically create new R objects based on expressions using assign:

library(tidyverse)
columns <-
  iris %>%
  colnames() %>%
  setdiff("Species")
columns
#> [1] "Sepal.Length" "Sepal.Width"  "Petal.Length" "Petal.Width"

columns %>%
  walk(~ {
    data <- iris %>% head(2) %>% select_at(c("Species", .x))
    assign(.x, data, envir = globalenv())
  })

# access created objects
Sepal.Width
#>   Species Sepal.Width
#> 1  setosa         3.5
#> 2  setosa         3.0
Sepal.Length
#>   Species Sepal.Length
#> 1  setosa          5.1
#> 2  setosa          4.9

^{Created on 2021-11-25 by the reprex package (v2.0.1)}

CodePudding user response：

Adding to danlooos answer, if you want a Base R solution you could just use a loop:

for (col in colnames(iris)[-1:-2]) {
    assign(col, iris[, c("Sepal.Length", "Sepal.Width", col)], envir = globalenv())
}

Or do the same thing and store the resulting data frames in a list, which I personally find somewhat cleaner:

new_frames <- list()
for (col in colnames(iris)[-1:-2]) {
    new_frames <- append(new_frames, list(iris[, c("Sepal.Length", "Sepal.Width", col)]))
}