CODEWITHSUNDEEP
rdataframe
CodeGuy
CodeGuy

Reputation: 28905

Data Frames in R - NA values

I'm trying to remove all the rows from my data frame in which the 3rd column is NA:

new.frame <- data[(!is.na(data$z)),]

But I'm getting an error. 

Warning message:
In is.na(data$z) : is.na() applied to non-(list or vector) of type 'NULL'

How can I accomplish this?

Upvotes: 1

Views: 6688

Answers (4)

RHelp
RHelp

Reputation: 835

You can also use na.omit function on the whole dataset.

Upvotes: 1

edgester
edgester

Reputation: 503

Try this:

 new.frame.nonull <- data[(!is.null(data$z)),]
 new.frame <- new.frame.nonull[(!is.na(new.frame.nonull$z)),]

Upvotes: 0

Chase
Chase

Reputation: 69241

There's also complete.cases() which may be easier to read. Using Dirk's data:

new.frame2 <- DF[complete.cases(DF) ,]

> all.equal(new.frame, new.frame2)
[1] TRUE

Upvotes: 3

Dirk is no longer here
Dirk is no longer here

Reputation: 368599

Reproducible examples, please. Here is one that works:

R> set.seed(42)
R> DF <- data.frame(a=rnorm(10), b=sample(LETTERS, 10, replace=TRUE), 
+>                  z=cumsum(runif(10)))
R> DF[c(2,4,6),"z"] = NA
R> DF
            a b        z
1   1.3709584 X 0.737596
2  -0.5646982 D       NA
3   0.3631284 Z 1.936759
4   0.6328626 Y       NA
5   0.4042683 C 2.625877
6  -0.1061245 N       NA
7   1.5115220 K 3.466127
8  -0.0946590 X 3.673786
9   2.0184237 L 4.580388
10 -0.0627141 V 5.192166
R> new.frame <- DF[(!is.na(DF$z)),]
R> new.frame
            a b        z
1   1.3709584 X 0.737596
3   0.3631284 Z 1.936759
5   0.4042683 C 2.625877
7   1.5115220 K 3.466127
8  -0.0946590 X 3.673786
9   2.0184237 L 4.580388
10 -0.0627141 V 5.192166
R>

Upvotes: 9

Related Questions