How to conditionally rename multiple columns of a dataframe using values from a sequence as new variable names?

0
votes

I would like to use the values of this sequence seq to rename the columns x1 to x10 of the data frame df.

the sequence - variable names to be used

var_names<-seq(10, 23.5, 1.5)

the data frame, where all variables beginning with x are to be renamed by the values of the sequence var_names

df = 
  data.frame( x1 = rlnorm(10000), x2 = rlnorm(10000), x3 = rlnorm(10000), x4 = rlnorm(10000), x5 = rlnorm(10000), x6 = rlnorm(10000), x7 = rlnorm(10000), x8 = rlnorm(10000), x9 = rlnorm(10000), x10 = rlnorm(10000))

my attempt

df_renamed <- apply(df, 2, function(x) data.table::setnames(df, old=colnames(x), new=var_names))

## Error in data.table::setnames(df, old = colnames(x), new = var_names) : 
##  'new' is not a character vector

expected output

In the output dataframe I want to replace the variable names x1 by 10, x2 by 11.5, x3 by 13, x4 by 14.5, x5 by 16, x6 by 17.5, x7 by 19, x8 by 20.5, x9 by 22, x10 by 23.5.

While this replacement of the variable names can be done individually using this and other approaches when there are only a few columns, I want a function that does the job efficiently with large datasets using directly the sequence var_names and taking advantage of the fact that all variables to be renamed begin with x.

1
r
Why don't use colnames(df) <- var_names ? - Clemsang
That gives 'Error in names(x) <- value : 'names' attribute [11] must be the same length as the vector [10]' - Krantz
It works for me - Clemsang
let me check again. thanks - Krantz
You should have just done data.table::setnames(df, var_names) - David Arenburg

1 Answers

2
votes

Perhaps colnames can do the job

colnames(df) <- var_names

Just a remark: It is not the best idea to have numbers as colnames, as it can be confusing and not easy to work with (for example you need to quote such names with ` `)