1

I have this vector: 

vector <- c(NA,1,rep(NA,24),1,NA,NA)
 [1] NA  1 NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA  1 NA NA

Essentially what I want to do is:

In excel the command is this: Calculation point = E40

=IF(B41=1,1,+E39+1)

In R it would be like: 

ifelse([+1 forward] == 1,1,[-1 behind]+1)

It is basically a counter which resets to 1 when a 1 is encountered.

This is desired output: 

> output
 [1]  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25  1  2  3  4

If we start in element 1, vector[1] then we are looking at vector[2] with the statement ifelse([+1 forward],1,1

in this case, this is true so we place 1... next we move up a element position. Now we in position 2.. vector[2]. The statement is now looking for a ifelse([+1 forward],1,1 in vector[3], if its not a 1, else its taking vector[1] and adding a +1. Next thing... at position vector[3], the statement is now looking for a ifelse([+1 forward],1,1 in vector[4], if its not a 1, else its taking vector[2] and adding a +1.. so a running counter... eventually we meet another 1, then the counter will reset to 1... and we begin counting again... until next 1.

making sense maybe?

here is how it works in excel

enter image description here

This is what I have tried: 

vector <- c(NA,1,rep(NA,24),1,NA,NA)
i=1
for( i in 1:length(vector)) {
  start <- ifelse(vector[i + 1]==1,1,vector[i - 1]+1)
}
start
Andrew Bannerman
  • 1,235
  • 2
  • 16
  • 36
  • `dplyr::lead` and `dplyr::lag`? – CPak Dec 10 '17 at 01:23
  • if i lead(vector) the very first element disappears and tacks on a NA at the end, when I need the first to stay in position 1 in order to add it to the ongoing calculation. – Andrew Bannerman Dec 10 '17 at 01:25
  • Sorry I can't really follow what you want with your example? If you could be more explicit about the outcome you want, I might be able to understand why `lead/lag` don't work for you – CPak Dec 10 '17 at 01:29
  • Just so that I understand correctly, if the value at the second position of the input vector was NA or a value different from 1, would the result still be the same? – Lamia Dec 10 '17 at 01:41
  • the 1 position would affect the output. If there is a 1, the counter will place a 1 in its place, ifelse vector element == 1,1, else previous number +1... So if it was NA,NA,NA,NA,1,NA... it would be 1,2,3,1,2 (its 1,2,3 as the ifelse when in position 4, its looking at position 5 for the ifelse 1=1) – Andrew Bannerman Dec 10 '17 at 01:46

3 Answers3

2

This reproduces your expected output.

x <- vector;
for (i in 1:length(x))
    if (x[i + 1] == 1 & !is.na(x[i + 1])) x[i] = 1 else x[i] = x[i - 1] + 1;
x;
# [1]  1  2  3  4  5  6  7  8  9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
#[26]  1  2  3  4

PS. vector is a very poor name for a variable in R.

Sample data

vector <- c(NA,1,rep(NA,24),1,NA,NA);
Maurits Evers
  • 49,617
  • 4
  • 47
  • 68
1

If I understand correctly what you're after, you could do this:

library(dplyr)
x = c(NA, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, NA, 1, 1, 0, 2, 5, 6, 1, 3)
y = (lead(x)!=1|is.na(lead(x)))
res = cumsum(y)-cummax((cumsum(y)-1)*!y)
[1] 1 2 3 4 5 6 1 2 3 4 5 6 1 1 2 3 4 5 1 2 3

This is a building up on one of the answers to this question.

Lamia
  • 3,845
  • 1
  • 12
  • 19
  • It is close! With my exact example... it places a 1,1,2,3,4,5,6 etc.. when it should be 1,2,3,4,5, right from the start... its pretty close though! I was trying a for loop off setting the i+1 and i-1 to represent looking forward and backwards – Andrew Bannerman Dec 10 '17 at 02:19
0

After a while figured it out with a for loop 

i=1
start <- NULL
for( i in 1:length(vector)) {
  start[i] <- ifelse(vector[i + 1]==2,1,start[i - 1]+1)
}
start

Data: 

 vector <- c(1,2,rep(1,24),2,1,1)
Andrew Bannerman
  • 1,235
  • 2
  • 16
  • 36
  • 1
    Did you not see my updated solution? Also, you don't need the `i=1` at the beginning; and it's better practice to pre-allocate memory for `start` by e.g. `start <- vector` or `start <- rep(0, length(vector))`. – Maurits Evers Dec 10 '17 at 02:47