Dynamically select data frame columns using $ and a vector of column names

Question

I wish to order a data frame based on different columns, one at a turn. I have a character vector with the relevant column names on which the order should be based:

parameter <- c("market_value_LOCAL", "ep", "book_price", "sales_price", "dividend_yield",

               "beta", "TOTAL_RATING_SCORE", "ENVIRONMENT", "SOCIAL", "GOVERNANCE")

I wish to loop over the names in parameter and dynamically select the column to be used to order my data:

Q1_R1000_parameter <- Q1_R1000[order(Q1_R1000$parameter[X]), ]

where X is 1:10 (because I have 10 items in parameter).

To make my example reproducible, consider the data set mtcars and some variable names stored in a character vector cols. When I try to select a variable from mtcars using a dynamic subset of cols, in a similar way as above (Q1_R1000$parameter[X]), the column is not selected:

cols <- c("cyl", "am")

mtcars$cols[1]

# NULL

Answer 1

To dynamically select data frame columns, you can use the [ (or [ [ if you want to extract only a single column as a vector).

For example:

set.seed(123)

df <- data.frame( A = sample(5,10,replace = TRUE) , B = sample(5,10,replace = TRUE) , C = sample(5,10,replace = TRUE) )

   A B C

1  3 5 2

2  3 3 1

3  2 3 3

4  2 1 4

5  3 4 1

6  5 1 3

7  4 1 5

8  1 5 4

9  2 3 2

10 3 2 5

To order using do.call:

sort_by <- c("C","A")

df[ do.call( order , df[ , match( sort_by , names(df) ) ]  ) , ]

Output:

   A B C

2  3 3 1

5  3 4 1

9  2 3 2

1  3 5 2

3  2 3 3

6  5 1 3

8  1 5 4

4  2 1 4

10 3 2 5

7  4 1 5

In the above code, we have used  'do.call' to call order. The second argument in do.call is a list of arguments that we pass to the first argument, in this case 'order'. Since a data.frame is really a list, we just subset the data.frame according to the columns we want to sort in, in that order.