CODEWITHSUNDEEP
r
Anton
Anton

Reputation: 4815

Indexing With A Logical Test

I want a variable containing all the rows of dataframe df where x is greater than y. But for some reason I keep getting this strange error.

> x <- c(5,6,7,8,9,10)
> y <- c(1,7,8,29,0,1)
> 
> df <- data.frame(x, y)
> 
> x.is.more <- df[,"x" > "y"]; x.is.more
data frame with 0 columns and 6 rows

The code above should the same results as x.is.more <- df[c(1,6),] but for some reason it isn't working.

Upvotes: 4

Views: 63

Answers (4)

zx8754
zx8754

Reputation: 56004

Try this:

#dummy dataframe
df <- data.frame(x=c(5,6,7,8,9,10),
                 y=c(1,7,8,29,0,1))

#find x>y
x.is.more <- df[df$x > df$y, ]

#output
x.is.more
#   x y
#1  5 1
#5  9 0
#6 10 1

Upvotes: 1

Anton
Anton

Reputation: 4815

Aha! I figured it out.

Instead of,

x.is.more <- df[,"x" > "y"]

I should have,

x.is.more <- df[x > y,]

Upvotes: 0

James
James

Reputation: 66834

subset is probably the most straightforward way of doing this:

subset(df, x>y)
   x y
1  5 1
5  9 0
6 10 1

Upvotes: 3

Jilber Urbina
Jilber Urbina

Reputation: 61154

You can use transform if you want to append the new variable to the existing data frame

> transform(df, x.is.more= x>y)
   x  y x.is.more
1  5  1      TRUE
2  6  7     FALSE
3  7  8     FALSE
4  8 29     FALSE
5  9  0      TRUE
6 10  1      TRUE

Use with if you just want a new variable

> x.is.more <- with(df, x>y)
> x.is.more
[1]  TRUE FALSE FALSE FALSE  TRUE  TRUE

If you want to subset then try

> df[ with(df, x>y), ]  # equivalent to df[ x.is.more, ]
   x y
1  5 1
5  9 0
6 10 1

Or directly use subset function as in @James answer

Upvotes: 2

Related Questions