Get the nth lagged value in grouped data in R

Asked Oct 11 '18 at 12:06

Active Oct 11 '18 at 12:06

Viewed 88 times

I have a data frame similar to this

mydf =  data_frame(letter = rep(c("a", "b", "c"), each =5), 
                 var1 = sample(1:25, 15, replace = TRUE))

# A tibble: 15 x 2
   letter  var1
   <chr>  <int>
 1 a         16
 2 a          9
 3 a          5
 4 a         14
 5 a          6
 6 b         13
 7 b          9
 8 b         20
 9 b         18
10 b          4
11 c         18
12 c         11
13 c          9
14 c          1
15 c         12

I know I can get the immediate value from the previous row with dplyr::lag(). However, I am trying to obtain a similar solution to obtain the third value before each observation. The expected result should look like this:

# A tibble: 15 x 3
# Groups:   letter [3]
   letter  var1  var2
   <chr>  <int> <dbl>
 1 a         16   NA 
 2 a          9   NA
 3 a          5   16
 4 a         14    9
 5 a          6    5
 6 b         13   NA 
 7 b          9   NA
 8 b         20   13

Thanks in advance

asked Oct 11 '18 at 12:06

Edu

4

Have you tried to use the `n` argument within `lag`? Like `lag(x, n=3)` to get value from 3 positions before. – AntoniosK Oct 11 '18 at 12:08
Why are rows 6 and 7 on var2 NA? – tmfmnk Oct 11 '18 at 12:13
@tmfmnk looks like he's grouping by `letter` column. – AntoniosK Oct 11 '18 at 12:21
@AntoniosK `n` refers to the number of positions to skip, so to get the third position it's actually `lag(x, n = 2)` – camille Oct 11 '18 at 12:42
@camille yes, you're right. – AntoniosK Oct 11 '18 at 12:43
That´s the one. Many thanks – Edu Oct 11 '18 at 13:04

Get the nth lagged value in grouped data in R

0 Answers0