Subset data frame based on multiple conditions

Question

enter image description here

I wish to filter a data frame based on conditions in several columns. For example, how can I delete rows if column A = B and Column E = 0.

It is not a good idea to alter Q like this -- older answers get invalidated and it brings confusion. Try asking new question in the future. — mbq, Nov 04 '11 at 09:05
related : http://stackoverflow.com/questions/6650510/remove-rows-from-data-frame-where-a-row-match-a-string — Joris Meys, Nov 04 '11 at 12:18

score 338 · Accepted Answer · answered Nov 04 '11 at 09:04

338

Logic index:

d<-d[!(d$A=="B" & d$E==0),]

answered Nov 04 '11 at 09:04

mbq

4

I tried so many complicated answers -- none worked. Your solution is simple and brilliant. – WGray Aug 06 '15 at 20:49
5

Actually, a simpler way of viewing it is: `foo.isolated – WGray Aug 06 '15 at 21:01
1

Why are NA's still included – Jørgen K. Kanters Dec 03 '16 at 13:13
NA's killed my two hours :D Note that it will also select NA's – Ioane Sharvadze Feb 24 '17 at 20:18

Tyler Rinker · Answer 2 · 2011-11-04T13:36:15.363

99

Subset is your safest and easiest answer.

subset(dataframe, A==B & E!=0)

Real data example with mtcars

subset(mtcars, cyl==6 & am!=0)

edited Nov 04 '11 at 13:36

answered Nov 04 '11 at 05:37

Tyler Rinker

1

a good second example would be to filter on a list of values. ie subset of mtcars for cyl not in c(100, 200, 500) – airstrike Nov 09 '15 at 04:16

score 3 · Answer 3 · answered Nov 04 '11 at 05:33

3

Use the which function:

A <- c('a','a','b','b','b')
B <- c(1,0,1,1,0)
d <- data.frame(A, B)

r <- with(d, which(B==0, arr.ind=TRUE))
newd <- d[-r, ]

answered Nov 04 '11 at 05:33

Manuel Ramón

5

-1. Don't use `which` for this. Change the condition to `B==2` and see if it gives the answer that you want. See, e.g., http://rwiki.sciviews.org/doku.php?id=tips:surprises:traps#negative_array_indices – Richie Cotton Nov 04 '11 at 10:47
Why not to use `which` from [Advanced R](http://adv-r.had.co.nz/Subsetting.html): "there are two important differences. First, when the logical vector contains `NA`, logical subsetting replaces these values by NA while `which()` drops these values. Second, `x[-which(y)]` is not equivalent to `x[!y]`: if `y` is all `FALSE`, `which(y)` will be `integer(0)` and `-integer(0)` is still `integer(0)`, so you’ll get no values, instead of all values. In general, avoid switching from logical to integer subsetting unless you want, for example, the first or last TRUE value." – filups21 Sep 30 '19 at 14:17

3 Answers3