Read csv file in R with currency column as numeric

Question

I'm trying to read into R a csv file that contains information on political contributions. From what I understand, the columns by default are imported as factors, but I need the the amount column ('CTRIB_AMT' in the dataset) to be imported as a numeric column so I can run a variety of functions that wouldn't work for factors. The column is formatted as a currency with a "$" as prefix.

I used a simple read command to import the file initially:

contribs <- read.csv('path/to/file')

And then tried to convert the CTRIB_AMT from currency to numeric:

as.numeric(as.character(sub("$","",contribs$CTRIB_AMT, fixed=TRUE)))

But that didn't work. The functions I'm trying to use for the CTRIB_AMT columns are:

vals<-sort(unique(dfr$CTRIB_AMT))
sums<-tapply( dfr$CTRIB_AMT, dfr$CTRIB_AMT, sum)
counts<-tapply( dfr$CTRIB_AMT, dfr$CTRIB_AMT, length)

See related question here.

Any thoughts on how to import file initially so column is numeric or how to convert it after importing?

The `sub` -> `as.character` -> `as.numeric` conversion should do the trick. What do you mean by "didn't work"? — Daniel Dickison, Sep 07 '11 at 17:25

score 18 · Accepted Answer · answered Sep 07 '11 at 17:48

18

I'm not sure how to read it in directly, but you can modify it once it's in:

> A <- read.csv("~/Desktop/data.csv")
> A
  id   desc price
1  0  apple $1.00
2  1 banana $2.25
3  2 grapes $1.97
> A$price <- as.numeric(sub("\\$","", A$price))
> A
  id   desc price
1  0  apple  1.00
2  1 banana  2.25
3  2 grapes  1.97
> str(A)
'data.frame':   3 obs. of  3 variables:
 $ id   : int  0 1 2
 $ desc : Factor w/ 3 levels "apple","banana",..: 1 2 3
 $ price: num  1 2.25 1.97

I think it might just have been a missing escape in your sub. $ indicates the end of a line in regular expressions. \$ is a dollar sign. But then you have to escape the escape...

answered Sep 07 '11 at 17:48

Zach

2,315
3
15
25

I initially thought there was a missing escape too, but `fixed=TRUE` takes care of that ... – Ben Bolker Sep 07 '11 at 17:53
10

+1 I also found `gsub( "[$,]", "",` . . . useful for figures like `"$1,234"`. – Neil Best Apr 09 '13 at 01:18
3

This should not be the accepted solution; it is inferior to other solutions which do the substitution at read-time, at read-time, return numeric, and thus avoid creating lots of unnecessary and unique strings, which could waste Mb/Gb on a large file, and also be slower. – smci Feb 21 '19 at 00:25

score 13 · Answer 2 · edited May 23 '17 at 12:25

Another way could be setting conversion using setAs.
It was used in two (similar) question:

For your needs:

setClass("Currency")
setAs("character", "Currency",
    function(from) as.numeric(sub("$","",from, fixed=TRUE)))

contribs <- read.csv("path/to/file", colClasses=c(CTRIB_AMT="Currency"))

score 5 · Answer 3 · answered Apr 07 '13 at 19:30

Yet another solution for a problem solved long time ago:

convertCurrency <- function(currency) {
  currency1 <- sub('$','',as.character(currency),fixed=TRUE)
  currency2 <- as.numeric(gsub('\\,','',as.character(currency1))) 
  currency2
}

contribs$CTRIB_AMT_NUM <- convertCurrency(contribs$CTRIB_AMT)

Aurèle · Answer 4 · 2018-07-16T15:26:05.570

Taking advantage of the powerful parsers the readr package offers out of the box:

my_parser <- function(col) {
  # Try first with parse_number that handles currencies automatically quite well
  res <- suppressWarnings(readr::parse_number(col))
  if (is.null(attr(res, "problems", exact = TRUE))) {
    res
  } else {
    # If parse_number fails, fall back on parse_guess
    readr::parse_guess(col)
    # Alternatively, we could simply return col without further parsing attempt
  }
}

library(dplyr)

name <- c('john','carl', 'hank')
salary <- c('$23,456.33','$45,677.43','$76,234.88')
emp_data <- data.frame(name,salary)

emp_data %>% 
  mutate(foo = "USD13.4",
         bar = "£37") %>% 
  mutate_all(my_parser)

#   name   salary  foo bar
# 1 john 23456.33 13.4  37
# 2 carl 45677.43 13.4  37
# 3 hank 76234.88 13.4  37

score 2 · Answer 5 · answered Sep 07 '11 at 17:28

Or use something like as.numeric(substr(as.character(contribs$CTRIB_AMT),2,20)) we know that there certainly won't be more than 20 characters.

Another thing to note is that you can remove the need to convert from a factor alltogether if you set stringsAsFactors=F in your call to read.csv()

Read csv file in R with currency column as numeric

5 Answers5

Linked

Related