CODEWITHSUNDEEP
rif-statementconditional-statements
joffie
joffie

Reputation: 215

R if statement with multiple conditions and multiple outcomes

I wonder if there exists a function for the following. I have a dataframe as following

test <- data.frame(replicate(2,sample(0:100,1000,rep=TRUE)))
cutoff.X1 <- 20
cutoff.X2 <- 60
test
    X1 X2
1   63 79
2   68 76
3   23 67
4   21 48
5   78 84
6   35 47
7   34 20
8   24 36
9   32 41
10  92 90
11  77 20
12  21 54
13  90 81
14  69 43
15  70 30
16  56 31
17   1 74
18 100 71
19  72 36
20  88 55

What I want is an added column stating X1, X2, none or both, depending on whether the row in X1 is above cutoff.X1, the vaule of X2 is above cutoff.X2, both are not, or both are above the cutoffs. I would know how to do it using multiple If statements, but since the real code is a lot of data, I wonder if there is a way without processing more data than necessary.

Upvotes: 2

Views: 2516

Answers (3)

Ian Campbell
Ian Campbell

Reputation: 24878

Here's a simple approach using some math and logical vectors. This leverages the fact that + TRUE evaluates to 1.

First, make a character vector with the choices. Then Test if X1 is greater than the cutoff. That will equal 1 when TRUE. Then test if X2 is greater than the cutoff and multiply by 2. Then, add those two numbers together and add 1. The total will equal 1 when neither, 2 when X1 is greater, 3 when X2 is greater, and 4 when both are.

Finally, subset your character vector using the integer vector you just created.

test$above <- c("neither","X1","X2","both")[((test$X1 > cutoff.X1) + ((test$X2 > cutoff.X2) * 2)) + 1]
head(test,10)
   X1 X2   above
1  64 51      X1
2  39 31      X1
3  24 14      X1
4  74 57      X1
5  67 91    both
6   7  6 neither
7  14 78      X2
8  74 92    both
9  18 93      X2
10 27 31      X1

If you don't like that, there's always dplyr::case_when, which is admittedly easier to read:

library(dplyr)
test$above <- case_when(test$X1 > cutoff.X1 & test$X2 > cutoff.X2 ~ "both",
                        test$X1 > cutoff.X1 ~ "X1",
                        test$X2 > cutoff.X2 ~ "X2",
                        TRUE ~ "neither")

Just remember that case_when evaluates each condition until one evaluates TRUE, so be sure to put the "both" condition first.

Upvotes: 3

Claudio Paladini
Claudio Paladini

Reputation: 1028

I think you are after an if_else() from dplyr:

test <- data.frame(replicate(2,sample(0:100,1000,rep=TRUE)))
cutoff.X1 <- 20
cutoff.X2 <- 60
test %>% 
  mutate(
    X3 = if_else(
      X1 > cutoff.X1,
      if_else(
        X2 > cutoff.X2,
        "both",
        "X1"
      ),
      if_else(
        X2 > cutoff.X2,
        "X2",
        "none"
      )
    )
  )

Result:

     X1  X2   X3
1     9   1 none
2    32  30   X1
3    30   7   X1
4    79  36   X1
5    70   0   X1
6     0  12 none
7    59  21   X1
8     5  38 none
9    57   4   X1
10   41  69 both
11   20  98   X2

Upvotes: 2

Terru_theTerror
Terru_theTerror

Reputation: 5017

You can use multiple if statements in this way:

test$Result<-ifelse(test$X1<cutoff.X1,"X1 under cutoff","")

One for every condition you want to check.

Upvotes: 1

Related Questions