Conditional Statements; creating new binary variable

Question

I have a pair of binary variables (1's and 0's), and my professor wants me to create a new binary variable that takes the value 1 if both of the previous variables have the value 1 (i.e., x,y=1) and takes the value zero otherwise.

How would I do this in R?

Thanks! JMC

Community · Accepted Answer · 2017-05-23 10:32:16Z

3

Here's one example with some sample data to play with:

set.seed(1)
A <- sample(0:1, 10, replace = TRUE)
B <- sample(0:1, 10, replace = TRUE)
A
#  [1] 0 0 1 1 0 1 1 1 1 0
B
#  [1] 0 0 1 0 1 0 1 1 0 1

as.numeric(A + B == 2)
#  [1] 0 0 1 0 0 0 1 1 0 0

as.numeric(rowSums(cbind(A, B)) == 2)
#  [1] 0 0 1 0 0 0 1 1 0 0

as.numeric(A == 1 & B == 1)
# [1] 0 0 1 0 0 0 1 1 0 0

Update (to introduce some more alternatives and share a link and a benchmark)

set.seed(1)
A <- sample(0:1, 1e7, replace = TRUE)
B <- sample(0:1, 1e7, replace = TRUE)

fun1 <- function() ifelse(A == 1 & B == 1, 1, 0)
fun2 <- function() as.numeric(A + B == 2)
fun3 <- function() as.numeric(A & B)
fun4 <- function() as.numeric(A == 1 & B == 1)
fun5 <- function() as.numeric(rowSums(cbind(A, B)) == 2)

library(microbenchmark)
microbenchmark(fun1(), fun2(), fun3(), fun4(), fun5(), times = 5)
# Unit: milliseconds
#   expr       min        lq    median        uq        max neval
# fun1() 4842.8559 4871.7072 5022.3525 5093.5932 10424.6589     5
# fun2()  220.8336  220.9867  226.1167  229.1225   472.4408     5
# fun3()  440.7427  445.9342  461.0114  462.6184   488.6627     5
# fun4()  604.1791  613.9284  630.4838  645.2146   682.4689     5
# fun5()  373.8088  373.8532  373.9460  435.0385  1084.6227     5

As can be seen, ifelse is indeed much slower than the other approaches mentioned here. See this SO question and answer for some more details about the efficiency of ifelse.

edited May 23, 2017 at 10:32

CommunityBot

11 silver badge

answered Dec 7, 2013 at 3:52

A5C1D2H2I1M1N2O1R2T1

194k31 gold badges417 silver badges497 bronze badges

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

9 Comments

user3076619 Over a year ago

Thanks! I wasn't familiar with the rowSums() function, which made the problems far easier.

gung - Reinstate Monica Over a year ago

It's interesting to me that you approach the issue in this way. It seems like things I've seen a number of times on SO. I think ifelse() is more intuitive, especially if someone has some basic programming before coming to R (not a criticism). Is there a simple reason you prefer these types of solutions?

A5C1D2H2I1M1N2O1R2T1 Over a year ago

@gung, because R is vectorized and these approaches are likely to be faster :-)

gung - Reinstate Monica Over a year ago

Isn't ifelse() vectorized? I thought it was. I know that if(){...} else {...} is not vectorized, & I try not to use it.

A5C1D2H2I1M1N2O1R2T1 Over a year ago

@gung, I'm out right now, but will update later with some benchmarks. I seem to remember a question about this too, so I'll try to track that down for you.

|

Collectives™ on Stack Overflow

Conditional Statements; creating new binary variable

1 Answer 1

Update (to introduce some more alternatives and share a link and a benchmark)

9 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Update (to introduce some more alternatives and share a link and a benchmark)

9 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related