CODEWITHSUNDEEP
rfunctionloops
Elong Chen
Elong Chen

Reputation: 99

continually multiply many columns to form a new variable

I recently want to continually multiply many columns and form a new variable in R.

I found that there is a prod function in R, which could provide the product of all elements in many vectors or matrices.

However, I do not want to get the product of all element. Instead, I want to get the product of each row.

I can do it with code with loop as following:

t <-
 data.frame(x1=round(round(runif(10),1)+1),
            x2=round(round(runif(10),1)+1))
for(i in 1:nrow(t)) {    t$y[i] <- prod(t[i,]) }

However, I wonder whether there is more efficient way or easy function, in that I can used like prod(t$1:t$2) and get the result I want.

Upvotes: 3

Views: 76

Answers (2)

LAP
LAP

Reputation: 6685

The base variant of @akrun's answer would be 

x <- matrix(1:20, ncol = 4)

     [,1] [,2] [,3] [,4]
[1,]    1    6   11   16
[2,]    2    7   12   17
[3,]    3    8   13   18
[4,]    4    9   14   19
[5,]    5   10   15   20

apply(x, 1, prod)
[1]  1056  2856  5616  9576 15000

In contrast to matrixStats::rowProd(), this also works on a data.frame.

x <- as.data.frame(matrix(1:20, ncol = 4))
x$new <- apply(x, 1, prod)

> x
  V1 V2 V3 V4   new
1  1  6 11 16  1056
2  2  7 12 17  2856
3  3  8 13 18  5616
4  4  9 14 19  9576
5  5 10 15 20 15000

Upvotes: 3

akrun
akrun

Reputation: 887028

A convenient function would be rowProds from matrixStats, but that would require a matrix

library(matrixStats)
rowProds(as.matrix(t1))
#[1] 2 1 4 4 2 4 1 2 4 1

If we need a base R option, which works on data.frame, then use

Reduce(`*`, t1)
#[1] 2 1 4 4 2 4 1 2 4 1

data

set.seed(24)
t1 <- data.frame(x1=round(round(runif(10),1)+1),x2=round(round(runif(10),1)+1))

Upvotes: 4

Related Questions