How to loop over column names that contain a variable in R?

Question

I have a dataframe df:

Shares    Price1    Price2    Price3
100       9         10        11
200       5         6         7
300       3         2         1

And I would like to loop over this dataframe and create three new columns equal to Shares x Price[i] where (i in 1:3). I tried the following code:

for (j in 1:3) {
  df$paste0("MktCap",j,sep="")<-df$Shares*df$paste0("Price",j,sep="")
}

But I get the error:

Error: attempt to apply non-function

I've looked here but it's not quite what I want as I would like my new column names to iterate.

what is your expected output

BENY
– BENY

2017-07-29 22:43:36 +00:00
Commented Jul 29, 2017 at 22:43 — BENY
– BENY, Commented Jul 29, 2017 at 22:43

BENY · Accepted Answer · 2017-07-29 22:47:52Z

3

Is this what you want ? also, check the link here http://www.statmethods.net/management/subset.html

for (j in 1:3) {
  df[,paste0("MktCap",j,sep="")]<-df$Shares*df[,paste0("Price",j,sep="")]
}


> df
  Shares Price1 Price2 Price3 MktCap1 MktCap2 MktCap3
1    100      9     10     11     900    1000    1100
2    200      5      6      7    1000    1200    1400
3    300      3      2      1     900     600     300

answered Jul 29, 2017 at 22:47

BENY

324k22 gold badges176 silver badges250 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Oriol Mirosa · Accepted Answer · 2017-07-29 23:19:30Z

1

@Wen's solution works, and if you have many price columns it would be the way to go. But I think using dplyr you get a much more expressive solution that is easier to read and understand:

library(dplyr)

df <- data.frame(Shares = c(100, 200, 300), Price1 = c(9, 5, 3), Price2 = c(10, 6, 2), Price3 = c(11, 7, 1))

(df <- df %>%
  mutate(MktCap1 = Shares * Price1,
         MktCap2 = Shares * Price2,
         MktCap3 = Shares * Price3))

  Shares Price1 Price2 Price3 MktCap1 MktCap2 MktCap3
1    100      9     10     11     900    1000    1100
2    200      5      6      7    1000    1200    1400
3    300      3      2      1     900     600     300

answered Jul 29, 2017 at 23:19

Oriol Mirosa

2,8361 gold badge16 silver badges15 bronze badges

Comments

Rich Pauloo · Accepted Answer · 2017-07-30 01:01:59Z

Consider the data frame, df:

df = tribble(
~Shares,    ~Price1,    ~Price2,    ~Price3,
100,        9,          10,         11,
200,        5,          6,          7,
300,        3,          2,          1
)

First approach - Terrible. Hard coding. This works, but you want a reproducible solution.

df$Value1 = df$Shares * df$Price1
df$Value2 = df$Shares * df$Price2
df$Value3 = df$Shares * df$Price3

Second approach - Better, but still not great. Subset orginal data frame for values, multiply by prices, assign colnames, merge the data together

stockPrice = df[,2:4]
stockValue = df$Shares * stockPrice
colnames(stockValue) = c(paste("value", seq(1:3), sep = ""))
cbind(df, stockValue)

Third (best) approach - define a function!

calculateValues = function(df){
  N = ncol(df)
  L = N-1
  stockPrice = df[,2:N]
  stockValue = df$Shares * stockPrice
  colnames(stockValue) = c(paste("value", seq(1:L), sep = ""))
  cbind(df, stockValue)
}

calculateValues(df)

This should output a new data frame with shares * values each time, named and everything! The only catch is that the first column of your df has to be named "Shares" every time.

Collectives™ on Stack Overflow

How to loop over column names that contain a variable in R?

3 Answers 3

Comments

Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related