Replace negative values by NA values

Question

I have positive, negative and NA values in a Table, I need to replace negative values by NA values. Positive and NA values should remain as they are in Table. My Data set is similar to the one below:

NO. q
1   NA
2   NA
3   -133.6105198
4   -119.6991209
5   28.84460104
6   66.05345087
7   84.7058947
8   -134.4522694
9   NA
10  NA
11  73.20465643
12  -69.90723514
13  NA
14  69.70833003
15  65.27859906

I tried this:

if (q>0) {
    q=NA
} else {
    q=q
}

data is like this: here "q" name of column NA NA -133.6105198 -119.6991209 28.84460104 66.05345087 84.7058947 -134.4522694 NA NA 73.20465643 -69.90723514 NA 69.70833003 65.27859906 — Tika Ram Gurung, Feb 24 '15 at 12:41
Hi and thanks for the question. Since you are a new user, you'd probably like to read the Tour and also either accept the answer that covers you the most (green tick next to each answer), or provide additional information if no answer covers you 100%. — codingEnthusiast, Feb 24 '15 at 13:23

score 12 · Answer 1 · edited Feb 10 '18 at 18:17

12

Or use replace:

> df$q2 <- replace(df$q, which(df$q < 0), NA)
> df
   NO.          q       q2
1    1         NA       NA
2    2         NA       NA
3    3 -133.61052       NA
4    4 -119.69912       NA
5    5   28.84460 28.84460
6    6   66.05345 66.05345
7    7   84.70589 84.70589
8    8 -134.45227       NA
9    9         NA       NA
10  10         NA       NA
11  11   73.20466 73.20466
12  12  -69.90724       NA
13  13         NA       NA
14  14   69.70833 69.70833
15  15   65.27860 65.27860

Or with data.table:

library(data.table)
setDT(df)[q < 0, q := NA]

Or with replace in a dplyr pipe:

library(dplyr)
df %>% mutate(q = replace(q, which(q<0), NA))

edited Feb 10 '18 at 18:17

jpgard

653
7
15

answered Feb 24 '15 at 12:49

talat

68,970
21
126
157

How do I adjust the dplyr solution that it applies to the whole dataset? – KC15 May 16 '22 at 06:11
`mutate(across(everything(), function(x){replace(x, which(x<0), NA)}))` – Tom Bishop Aug 12 '22 at 09:55

score 9 · Answer 2 · answered Feb 24 '15 at 12:52

9

You could try this:

sample <- c(1, -2, NA)
sample[sample < 0] <- NA
sample
[1]  1 NA NA

Or if you're using a data.frame (suppose it's called df):

df$q[df$q < 0] <- NA

answered Feb 24 '15 at 12:52

codingEnthusiast

3,800
2
25
37

akrun · Answer 3 · 2015-02-24T13:20:36.533

6

You could try

df1$q1 <- NA^(df1$q <0) * df1$q
df1
#   NO.          q       q1
#1    1         NA       NA
#2    2         NA       NA
#3    3 -133.61052       NA
#4    4 -119.69912       NA
#5    5   28.84460 28.84460
#6    6   66.05345 66.05345
#7    7   84.70589 84.70589
#8    8 -134.45227       NA
#9    9         NA       NA
#10  10         NA       NA
#11  11   73.20466 73.20466
#12  12  -69.90724       NA 
#13  13         NA       NA
#14  14   69.70833 69.70833
#15  15   65.27860 65.27860

Or use ifelse

 with(df1, ifelse(q < 0, NA, q))

Or

 is.na(df1$q) <- df1$q < 0

edited Feb 24 '15 at 13:20

answered Feb 24 '15 at 12:42

akrun

874,273
37
540
662

q[which(q < 0)] <- NA – Tika Ram Gurung Feb 24 '15 at 14:28
@TikaRamGurung You can use any of the solutions to get the result, `which` is not necesssary `q[q<0] <- NA` would also work as shown by naltipar – akrun Feb 24 '15 at 16:12

score 1 · Answer 4 · answered Feb 11 '18 at 02:05

1

Another way of accomplishing the same thing is (now I see this is ALMOST the same as another answer by akrun, sorry for that)

daf$q = ifelse(daf$q < 0, NA_real_, daf$q)

answered Feb 11 '18 at 02:05

catastrophic-failure

3,759
1
24
43

Replace negative values by NA values

4 Answers4